Oil depot state low-cost monitoring method and system based on Internet of Things

Abstract

The invention provides an oil depot state low-cost monitoring method and system based on the Internet of Things, and relates to the technical field of oil depot detection, and the method comprises the steps: deploying a sensor group at the position of an oil depot storage tank, and collecting original monitoring data; preprocessing the collected original monitoring data, and then uploading the data to a cloud server through an Internet of Things gateway; based on the received data, the cloud server performs inversion calculation on the real-time liquid level height of the storage tank through an intelligent algorithm, and performs calibration calculation on the volume of the special-shaped storage tank; state fusion analysis and abnormity diagnosis are carried out based on the real-time liquid level height, the volume data and the multi-source environment data; and triggering graded early warning according to a preset rule based on results of analysis and abnormality diagnosis. According to the method provided by the invention, hardware deployment cost is revolutionarily reduced, and all-around efficiency leap from manual regular inspection to unmanned real-time monitoring, from single parameter measurement to multi-source fusion diagnosis, and from passive reading to active early warning is realized.

Description

Technical Field

[0001] This invention relates to the field of oil depot monitoring technology, specifically to a low-cost method and system for monitoring the status of oil depots based on the Internet of Things. Background Technology

[0002] Monitoring the liquid level and condition of oil depot storage tanks is a core aspect of ensuring energy storage safety and operational efficiency. Currently, the mainstream technologies in this field mainly include the following categories: First, direct-reading level gauges or float-type level gauges based on mechanical principles, which rely on physical structures for direct indication; second, non-contact ranging systems based on waves (such as ultrasonic waves, microwave radar) or light (such as lasers), which calculate the liquid level by measuring the signal propagation time; and third, intelligent solutions based on visual image analysis or high-precision industrial radar that have emerged in recent years. While these technologies can achieve a certain level of measurement accuracy, they all constitute the current technological foundation for oil depot monitoring.

[0003] However, the aforementioned existing technologies generally suffer from significant drawbacks: First, in terms of economy and scalability, high-precision radar or laser systems are extremely expensive to purchase and install, require a high level of professional maintenance, and are difficult to deploy comprehensively and densely in large oil depots. Second, in terms of functionality and reliability, mechanical instruments typically cannot perform electronic real-time monitoring and remote data transmission, while simple ranging systems often only provide a single parameter of liquid level, lacking multi-dimensional perception capabilities regarding tank status and environmental safety. Finally, in terms of operational models, existing technological systems cannot completely replace manual inspections, resulting in problems such as low efficiency, delayed response, and personnel safety risks.

[0004] Therefore, there is an urgent need for a low-cost, flexible, and integrated solution that can achieve truly unmanned intelligent monitoring. Summary of the Invention

[0005] To address the problems in related technologies, this invention provides a low-cost method and system for monitoring the status of oil depots based on the Internet of Things.

[0006] To achieve the above objectives, the technical solution adopted by the present invention includes: According to a first aspect of the present invention, a low-cost method for monitoring the status of oil depots based on the Internet of Things is provided, comprising: Step S1: Deploy sensor arrays at the location of the oil depot storage tanks and collect raw monitoring data; Step S2: Preprocess the collected raw monitoring data and then upload it to the cloud server through the IoT gateway; Step S3: Based on the received data, the cloud server uses intelligent algorithms to calculate the real-time liquid level of the storage tank and performs calibration calculations on the volume of the irregularly shaped storage tank. Step S4: Based on real-time liquid level height, volume data, and multi-source environmental data, perform state fusion analysis and anomaly diagnosis; Step S5: Based on the analysis and anomaly diagnosis results, trigger a tiered early warning according to preset rules.

[0007] Optionally, the sensor group includes a pressure sensor deployed at the bottom of the tank, a gas pressure sensor deployed at the top of the tank, and a temperature and humidity sensor deployed in the tank area environment; The raw monitoring data includes static pressure data measured by the pressure sensor, atmospheric pressure data measured by the air pressure sensor, and ambient temperature data measured by the temperature and humidity sensor.

[0008] Optionally, step S2 specifically includes: The static pressure data is compensated for zero-point drift based on real-time ambient temperature to obtain the compensated pressure value; The compensated data is then subjected to sliding window filtering and outlier removal based on statistical criteria. The processed data is packaged with the corresponding environmental data and device identification information, and then transmitted to the cloud server via a narrowband IoT network.

[0009] Optionally, in step S3, an improved hydrostatic level inversion model is used to calculate the real-time liquid level height H, and the calculation formula is as follows: In the formula, The compensation pressure value is obtained after zero-point drift compensation. The measured atmospheric pressure at the top of the tank. The density of the oil after compensation at the current temperature T. It is the acceleration due to gravity. This is a correction factor for the tank geometry related to the current liquid level. This is for compensation of thermal expansion and contraction of the tank. This is the dynamic disturbance compensation term estimated based on the Kalman filter.

[0010] Optionally, in step S3, for irregularly shaped storage tanks, the volume is calibrated using a height-volume polynomial fitting model, the expression of which is: In the formula, Let h be the liquid volume corresponding to the liquid level height. Let be the total volume of the storage tank, and n be a preset parameter selected based on the complexity of the tank's shape. k is the index of the summation term. The coefficients are the polynomial coefficients obtained by fitting using the least squares method. The maximum height of the storage tank; for irregularly shaped storage tanks, the tank geometry correction factor. Based on the derivative relationship of this model Dynamic calculation, in which, This is the reference area of ​​the bottom of the irregularly shaped storage tank.

[0011] Optionally, step S4 specifically includes: Construct an anomaly diagnosis model that integrates pressure time-series characteristics, liquid level change rate, and environmental sensor readings; When the storage tank is determined to be in a static state, the pressure time series data is analyzed. If the fit conforms to the exponential decay law, then... And attenuation constant If the threshold is exceeded, a suspected leak warning will be triggered.

[0012] Optionally, step S5 specifically includes: Real-time display of liquid level, inventory, environmental risk index, and equipment health status of each storage tank; A three-level early warning mechanism of "attention, warning, and severity" is set up, and the corresponding level of early warning information is automatically triggered based on the output of the anomaly diagnosis model.

[0013] Optionally, the IoT-based low-cost monitoring method for oil depot status further includes: Step S6: Perform end-to-end encryption on the data transmission link and perform two-way certificate authentication on the IoT gateway device.

[0014] Optionally, the IoT-based low-cost monitoring method for oil depot status further includes: Step S7: The cloud server periodically retrains the inversion model and anomaly diagnosis model in the intelligent algorithm using newly added monitoring data.

[0015] According to a second aspect of the present invention, a low-cost oil depot status monitoring system based on the Internet of Things is also provided for implementing the method described in any of the technical solutions of the first aspect of the present invention, comprising: The sensing and acquisition layer includes several pressure sensors, air pressure sensors, temperature and humidity sensors deployed on the tank body and tank area, as well as an IoT gateway for aggregating sensor data. The network transmission layer employs a narrowband IoT communication module to securely transmit data from the sensing and acquisition layer to the cloud. The cloud-based intelligent processing layer includes: The data access and storage module is used to receive and store monitoring data from the network transport layer; The liquid level inversion and volume calibration calculation module is used to perform real-time liquid level height and volume data calculation; The multi-source data fusion analysis and anomaly diagnosis module is used to perform state fusion analysis and anomaly diagnosis based on real-time liquid level height, volume data and multi-source environmental data. The visualization and early warning service module is used to trigger tiered early warnings based on the results of analysis and anomaly diagnosis, according to preset rules. The application presentation layer provides users with a visual interface to display monitoring results and receive early warning information.

[0016] Beneficial effects: 1. Through the above technical solution, firstly, through steps S1 and S2 of the present invention, unmanned and remote monitoring can be achieved. Specifically, by deploying sensor groups and uploading data using an IoT gateway, the method of the present invention, for the first time, continuously and automatically converts the physical state (pressure, environmental parameters) of key points in the oil depot into remotely accessible digital signals. Compared with existing technologies, the present invention directly replaces manual meter reading or local data collection operations that must be performed by personnel on-site. It can completely eliminate the personal safety risks of manual inspections in high-risk environments and improve the data acquisition frequency from a manual cycle of several hours / day to a real-time or near-real-time level at the second / minute level, laying the foundation for dynamic monitoring.

[0017] Secondly, step S3 of this invention enables a breakthrough in cost-effective monitoring. Specifically, the liquid level and volume are calculated via intelligent algorithm on a cloud server. Instead of directly measuring the liquid level, the liquid level is indirectly and accurately determined by measuring other readily available and lower-cost physical quantities (such as pressure) through an algorithm model. The physical basis of this calculation lies in the definite theoretical relationship between the tank liquid level and the static pressure at the bottom of the tank. The core task of the intelligent algorithm is to overcome practical interferences such as temperature changes, tank shape, and oil density fluctuations, and to faithfully reconstruct the liquid level value from pressure data. This makes it possible to replace expensive dedicated ranging instruments (such as radar and laser) with extremely low-cost pressure sensors, thereby achieving a significant reduction in system hardware costs at the source.

[0018] Third, through step S4 of this invention, automated diagnosis can be achieved. Specifically, the state fusion analysis and anomaly diagnosis in step S4 go beyond the initial stage of traditional monitoring methods that only provide readings (such as the current liquid level), and enter the interpretation stage. By fusing and analyzing real-time liquid level, volume data, and multi-source environmental data, the system can automatically diagnose anomalies. Compared with the prior art, the method of this invention can not only make qualitative judgments (for example, informing the user that the liquid level is dropping), but also make preliminary judgments (whether this drop is normal consumption or suspected leakage) by combining data such as valve status, historical trends, and ambient temperature. In this way, early warning and preliminary decision support capabilities can be provided, shifting accident handling from post-event response to in-event or even pre-event intervention, greatly improving the initiative and effectiveness of safety management.

[0019] Overall, the method of this invention, through a complete cloud-based intelligent processing architecture, with innovative indirect measurement principles at its core, revolutionarily reduces hardware deployment costs while achieving a comprehensive leap in efficiency, from manual periodic inspections to unmanned real-time monitoring, from single-parameter measurement to multi-source fusion diagnosis, and from passive readings to proactive early warnings.

[0020] 2. Other beneficial effects or advantages of the present invention will be described in detail in the specific embodiments. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] in: Figure 1 This is a flowchart illustrating the steps of a low-cost oil depot status monitoring method based on the Internet of Things, provided by an exemplary embodiment of the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0024] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0025] Furthermore, the terms "comprising" and "having," and any variations thereof, used in the description of this invention are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices. It should also be noted that in embodiments of this invention, the words "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in embodiments of this invention should not be construed as preferred or advantageous over other embodiments or designs. Specifically, the use of the words "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0026] To facilitate a clearer and more accurate understanding of the technical solutions of this invention by those skilled in the art, the existing related technologies and their technical problems will be described in more detail below.

[0027] I. Specific Examples of Existing Technical Solutions Existing technical solutions can be summarized into the following three categories, each with a clear application form and representative products: 1. Mechanical contact measurement. Typical equipment includes: magnetostrictive level gauges and steel strip float level gauges.

[0028] For example, a magnetostrictive level gauge is installed on the top of a 10,000 cubic meter vertical domed diesel storage tank. Its core component is a probe rod that penetrates vertically into the tank. A waveguide wire inside the rod interacts with a float suspended on the liquid surface. The float's position is determined by measuring the pulse propagation time, thus calculating the liquid level. Simultaneously, this instrument typically integrates multi-point temperature sensors (temperature measuring bulbs) for average temperature calculation. This is a direct-contact, point-level measurement method. It offers high accuracy (up to ±1 mm), but is a single-point, single-parameter measuring instrument.

[0029] 2. Non-contact wave / light ranging. Typical equipment includes: frequency modulated continuous wave (FMCW) radar level gauges and lidar (LiDAR) scanning systems.

[0030] For example, in a horizontal pressure tank storing volatile chemical raw materials, due to the presence of volatile gases and pressure inside, an FMCW radar level gauge with a horn-mouth antenna is often used. The instrument is installed on the top of the tank, emitting microwave signals towards the liquid surface and receiving the echoes; the distance is calculated by determining the frequency difference. Another more advanced application is in large crude oil storage areas, where lidar scanners mounted on inspection tracks or towers periodically scan the exterior of multiple tanks, retrieving the liquid level from the point cloud data. This is a non-contact, indirect measurement method. It avoids direct contact with the medium and is suitable for harsh operating conditions. Laser scanning systems further enable regional area measurements, but are essentially still geometric distance measurements.

[0031] 3. Visual analysis and fusion perception. Typical equipment includes: high-definition industrial cameras and image processing units, and digital twin systems based on 3D reality modeling (BIM).

[0032] For example, in some modern oil depots pursuing high levels of automation, arrays of smart cameras with protective casings are deployed around the tank area. These cameras continuously photograph the exterior of the tanks, using machine vision algorithms to identify pre-set scale markings on the tank walls or reflective spots on the liquid surface, thereby estimating the liquid level. Furthermore, by integrating drone oblique photogrammetry, a 3D model (BIM) of the entire tank area is constructed, visually integrating various sensor data within the model. This is an information fusion and visualization solution. It begins to attempt to integrate multi-source information and provide an intuitive display interface, but its core liquid level sensing often relies on the recognition of traditional markings (such as scales) or a combination with other point sensors.

[0033] In practical engineering applications, the above-mentioned technologies have revealed the following multi-layered limitations: First, there are bottlenecks in terms of economic efficiency and scalability.

[0034] First, the purchase cost of a high-performance FMCW radar level gauge or magnetostrictive level gauge typically ranges from RMB 10,000 to 50,000. A fixed lidar system for tank area scanning can cost hundreds of thousands of RMB. To achieve full coverage of a medium-sized oil depot with 50 tanks, the initial hardware investment for level monitoring alone could exceed one million RMB, constituting a significant capital barrier.

[0035] Secondly, radar level gauges require specialized drilling, flange connections, and calibration for installation; mechanical instruments have moving parts (such as floats and gears), which pose risks of jamming and wear during long-term operation, necessitating periodic tank shutdowns for maintenance. Laser scanning systems have extremely high requirements for the stability of the mounting base; even minor vibrations can cause measurement failure. These factors increase the overall lifecycle maintenance costs and hinder the rapid deployment of the technology in large-scale, aging tank renovation projects.

[0036] Second, limited functionality and data silos.

[0037] First, whether mechanical or radar-based, their core function is to measure liquid level or single-point temperature. They cannot directly and cost-effectively acquire critical safety and health parameters such as the tank's micro-deformation (e.g., stress changes), early leakage of combustible gas concentration in the tank area environment, or the operating status of pumps and valves. The safety status of oil depots requires monitoring using a separate sensor network (e.g., gas detectors, video surveillance), resulting in system fragmentation and the inability to link and analyze data.

[0038] Secondly, existing technologies primarily provide "what" (the current liquid level value), but struggle to answer "why" and "what will happen." For example, they cannot easily distinguish between a normal drop in liquid level and an abnormal drop caused by a minor leak, nor can they provide timely warnings of settlement trends in the tank foundation. This falls under the category of "monitoring" rather than "diagnosis" or "prediction."

[0039] Third, insufficient level of intelligence and reliance on manual labor.

[0040] First, mechanical instruments require manual on-site meter reading; while automated instruments can upload data, their own health status (such as antenna scaling, lens contamination, and mechanical jamming) still requires manual on-site inspection. Critical safety conditions such as external tank corrosion, loose bolts, and minor foundation changes rely entirely on visual inspection by patrol personnel. Therefore, regular manual inspections in high-risk environments cannot be omitted, and personnel safety risks and labor costs remain constant.

[0041] Secondly, manual inspection cycles are typically on an hourly or daily basis, resulting in slow response to rapidly evolving incidents (such as rapid leaks). Even automatic instruments that upload data in real time usually rely on simple threshold judgments (such as exceeding liquid level limits) for alarms, lacking early warning capabilities based on multi-parameter fusion and time-series pattern recognition. Alarms are often triggered only when the incident is already quite obvious, thus missing the opportunity to take proactive measures.

[0042] In summary, existing technologies have systemic shortcomings in terms of economy, functional integration, and intelligence, resulting in oil depot safety monitoring facing long-term dilemmas such as "incomplete coverage, shallow perception, slow early warning, and excessively high costs."

[0043] In view of this, the present invention provides a novel solution: a low-cost method for monitoring the status of oil depots based on the Internet of Things (IoT). The technical concept of this invention lies in abandoning the traditional approach of directly measuring liquid level and instead adopting a new paradigm of "indirect sensing-cloud-based computation." Specifically, the method of this invention does not rely on expensive dedicated ranging instruments. Instead, it deploys extremely inexpensive general-purpose pressure sensors at the bottom of the storage tank, and uses the IoT to upload basic physical quantities such as pressure and temperature to the cloud. In the cloud, an intelligent algorithm integrating fluid statics, geometric calibration, and dynamic compensation is used to convert pressure differences into high-precision liquid level and volume information. Furthermore, it integrates multi-source environmental data to construct a low-cost, high-dimensional status perception system that integrates real-time monitoring, early leak diagnosis, and equipment health warning. The essence of this invention is to transfer hardware costs and complexity to infinitely scalable cloud-based software intelligence, thereby revolutionarily reducing costs while achieving a fundamental functional leap from "single-parameter measurement" to "comprehensive status monitoring and diagnosis."

[0044] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings.

[0045] like Figure 1 As shown, this embodiment provides a low-cost method for monitoring the status of oil depots based on the Internet of Things (IoT) according to a first aspect of the present invention, comprising: Step S1: Deploy sensor arrays at the location of the oil depot storage tanks and collect raw monitoring data; Step S2: Preprocess the collected raw monitoring data and then upload it to the cloud server through the IoT gateway; Step S3: Based on the received data, the cloud server uses intelligent algorithms to calculate the real-time liquid level of the storage tank and performs calibration calculations on the volume of the irregularly shaped storage tank. Step S4: Based on real-time liquid level height, volume data, and multi-source environmental data, perform state fusion analysis and anomaly diagnosis; Step S5: Based on the analysis and anomaly diagnosis results, trigger a tiered early warning according to preset rules.

[0046] Through the above technical solution, firstly, steps S1 and S2 of this invention enable unmanned and remote monitoring. Specifically, by deploying sensor groups and uploading data using an IoT gateway, the method of this invention, for the first time, continuously and automatically converts the physical state (pressure, environmental parameters) of key points in an oil depot into remotely accessible digital signals. Compared with existing technologies, this invention directly replaces manual meter reading or local data collection operations that must be performed on-site by personnel. It can completely eliminate the personal safety risks of manual inspections in high-risk environments and improve the data acquisition frequency from a manual cycle of several hours / day to a real-time or near-real-time level at the second / minute level, laying the foundation for dynamic monitoring.

[0047] Secondly, step S3 of this invention enables a breakthrough in cost-effective monitoring. Specifically, the liquid level and volume are calculated via intelligent algorithm on a cloud server. Instead of directly measuring the liquid level, the liquid level is indirectly and accurately determined by measuring other readily available and lower-cost physical quantities (such as pressure) through an algorithm model. The physical basis of this calculation lies in the definite theoretical relationship between the tank liquid level and the static pressure at the bottom of the tank. The core task of the intelligent algorithm is to overcome practical interferences such as temperature changes, tank shape, and oil density fluctuations, and to faithfully reconstruct the liquid level value from pressure data. This makes it possible to replace expensive dedicated ranging instruments (such as radar and laser) with extremely low-cost pressure sensors, thereby achieving a significant reduction in system hardware costs at the source.

[0048] Third, through step S4 of this invention, automated diagnosis can be achieved. Specifically, the state fusion analysis and anomaly diagnosis in step S4 go beyond the initial stage of traditional monitoring methods that only provide readings (such as the current liquid level), and enter the interpretation stage. By fusing and analyzing real-time liquid level, volume data, and multi-source environmental data, the system can automatically diagnose anomalies. Compared with the prior art, the method of this invention can not only make qualitative judgments (for example, informing the user that the liquid level is dropping), but also make preliminary judgments (whether this drop is normal consumption or suspected leakage) by combining data such as valve status, historical trends, and ambient temperature. In this way, early warning and preliminary decision support capabilities can be provided, shifting accident handling from post-event response to in-event or even pre-event intervention, greatly improving the initiative and effectiveness of safety management.

[0049] Overall, the method of this invention, through a complete cloud-based intelligent processing architecture, with innovative indirect measurement principles at its core, revolutionarily reduces hardware deployment costs while achieving a comprehensive leap in efficiency, from manual periodic inspections to unmanned real-time monitoring, from single-parameter measurement to multi-source fusion diagnosis, and from passive readings to proactive early warnings.

[0050] The present invention will be further described below with reference to an exemplary embodiment.

[0051] This embodiment uses a vertical domed aviation kerosene storage tank at an international airport oil depot as the monitoring object.

[0052] Step 1: Sensor Deployment and Data Acquisition.

[0053] 1. Equipment deployment: Pressure sensor: An industrial-grade diffused silicon pressure transmitter with a range of 0-1MPa and an accuracy of 0.075%FS is selected and installed on the pipe near the drain outlet at the bottom of the storage tank to measure the sum of the liquid static pressure and the gas phase pressure inside the tank.

[0054] Pressure sensor: A high-stability absolute pressure sensor is selected and installed next to the vent on the top of the tank to measure the actual absolute pressure of the gas phase space inside the tank in real time.

[0055] Temperature and humidity sensor: Deployed on a weather pole outside the fire dike in the tank area to monitor the ambient temperature and humidity.

[0056] IoT Gateway: All sensor signals are connected to an intrinsically safe NB-IoT gateway located in an explosion-proof enclosure on site.

[0057] 2. Key environmental parameter settings: The oil depot is located at an average altitude of 500 meters, with an average annual atmospheric pressure of approximately 95.5 kPa (95,500 Pa), significantly lower than standard atmospheric pressure (101.325 kPa). In this embodiment, the measured values ​​of the tank top pressure sensor are set to fluctuate around this benchmark.

[0058] Ambient temperature: The average annual temperature in this region is approximately 16℃. Considering seasonal variations, this embodiment will simulate two typical working conditions for comparative demonstration: Operating Condition A (Spring and Autumn Normal Temperature): Ambient Temperature =15℃.

[0059] Operating Condition B (Winter Low Temperature): Ambient Temperature =5℃.

[0060] Tank parameters: Diameter 40 meters, tank wall height 15 meters, total volume Approximately 18,850 cubic meters. It is a regular vertical cylinder, therefore the shape correction factor is... ≡1.

[0061] Step 2: Data preprocessing and transmission.

[0062] The IoT gateway collects raw data at a frequency of 0.5Hz. Assume that at 10:00 AM on a winter day (condition B), a set of data is collected: the raw value of the tank bottom pressure sensor. =1045800Pa, tank top pressure sensor value =96000Pa, ambient temperature =5℃, average oil temperature ≈8℃.

[0063] Gateway performs preprocessing: 1. Temperature compensation: Zero-point temperature drift compensation is performed on the pressure sensor (assuming temperature drift coefficient α = 0.0005 / ℃).

[0064] =1045800×[1+0.0005×(5-20)]=1045800×0.9925≈1037955Pa.

[0065] 2. Filtering: Sliding window mid-value filtering is used.

[0066] 3. Package and upload: [This likely refers to a specific process or method, but without further context, it's difficult to translate accurately.] , , , The data will be encrypted and then uploaded to the cloud.

[0067] Step 3: Cloud-based intelligent inversion calculation.

[0068] After receiving the data, the cloud initiates the core algorithm.

[0069] 1. Density compensation calculation: Standard density of aviation kerosene =780kg / m³ (at 20℃), coefficient of thermal expansion β=0.0008 / ℃.

[0070] Operating condition A (oil temperature 15℃): ρ(15) = 780 / [1 + 0.0008 × (15 - 20)] ≈ 783.1 kg / m³; Operating condition B (oil temperature 8℃): ρ(8) = 780 / [1 + 0.0008 × (8 - 20)] ≈ 787.6 kg / m³.

[0071] It is evident that the density of oil increases significantly at low temperatures, which is a key factor that must be compensated for.

[0072] 2. Core liquid level inversion calculation: The liquid level height H is determined using an iterative method. The formula is: For vertical cylindrical tanks and We'll ignore this for now. =1.

[0073] First, calculate the measured static pressure difference: =1037955-96000=941955Pa; this It includes the static pressure of the oil column and possibly a small amount of gas phase pressure difference.

[0074] Iterative solution (taking working condition B as an example): A. First guess: Assumption =12.0m; B. Calculate the theoretical static pressure of the oil column: = =787.6×9.80665×12.0≈92685Pa; C. Comparison and Adjustment: The theoretical static pressure (92685 Pa) is much smaller than the measured static pressure difference (941955 Pa), indicating that the initial guess was significantly underestimated. This suggests... The value may contain a large amount of internal gas pressure that was not included in the subtraction (i.e., the internal gas pressure may be higher than the value of the external gas pressure). (Measurement points). This is a common and complex situation in actual engineering.

[0075] D. Intelligent Algorithm Correction: The cloud-based algorithm will identify this inconsistency and invoke redundant logic. One approach is to utilize... For reference, the actual pressure of the gas phase space inside the tank was estimated by combining the tank sealing model. Assume the algorithm estimates ≈1000000Pa.

[0076] E. Recalculate the effective static pressure difference: =1037955-1000000=37955Pa; F. Iterate again to solve for the liquid level: New speculation =5.0m; New theory of static pressure =787.6×9.80665×5.0≈38620Pa; The theoretical value is very close to the effective static pressure difference (37955Pa), and the algorithm converges.

[0077] G. Final Output: After dynamic compensation ( After fine-tuning, the cloud system determined the current liquid level H = 4.92 meters.

[0078] 3. Volume calculation: For vertical cylindrical tanks, the volume and liquid level have a linear relationship.

[0079] =18850×(4.92 / 15)≈6183 cubic meters.

[0080] Step 4: State fusion analysis and anomaly diagnosis.

[0081] The cloud continuously analyzes time-series data. A comparison of two operating conditions highlights the importance of environmental compensation. If the local low pressure in the region (95.5 kPa vs 101.3 kPa) is not taken into account, and the calculation is performed directly using standard atmospheric pressure, a systematic error of about 5.7 kPa will be introduced, resulting in a deviation of about 0.7 meters in the liquid level inversion.

[0082] If the effect of temperature on density is not considered, using the standard density (at 20℃) for calculation in winter (8℃) will result in a density deviation of approximately 7.6 kg / m³, which will also lead to significant errors in liquid level calculation.

[0083] The system monitors The long-term stability is used to diagnose anomalies. For example, during periods of no work, if... If a downward trend persists for several hours and exceeds the noise range, the algorithm will fit an exponential decay curve. If the fitted attenuation constant Greater than the threshold (e.g.) The system will trigger a "suspected micro-leak" warning.

[0084] Step 5: Visualization and Early Warning.

[0085] In the cloud-based control panel, maintenance personnel can see: Real-time data: Liquid level 4.92 m, stock 6183 The tank pressure is 1.000 MPa and the oil temperature is 8℃.

[0086] Environmental context: The interface prominently displays "Local atmospheric pressure: 96.0 kPa", and all calculations have been automatically compensated based on this.

[0087] Warning Status: The current system status is "Normal" (green). If a micro-leakage warning is triggered, it will be upgraded to "Attention" (yellow), and a message will be pushed to the responsible person.

[0088] System security and optimization: Security Mechanism: Intrinsically safe isolation is used from the sensor to the gateway. Data is encrypted with AES-128 in the gateway and transmitted to the cloud via the operator's NB-IoT network (with built-in link encryption). The cloud access end uses the TLS1.3 protocol and two-way certificate authentication.

[0089] Model optimization: The cloud-based algorithm automatically utilizes historical data from the past three months for the region each quarter (covering the full range from low to high temperatures) to re-fit the local atmospheric pressure-seasonal relationship model and the oil density-temperature curve model. This allows the system to adapt to the unique climate cycle of the region, continuously improving monitoring accuracy.

[0090] The present invention will now be described in conjunction with another embodiment.

[0091] This embodiment uses a horizontal cylindrical diesel storage tank at an international airport oil depot as the monitoring object. The cross-sectional area of ​​the horizontal tank changes non-linearly with the liquid level, making it a typical irregularly shaped tank, which can fully demonstrate the shape correction coefficient in the claims of this invention. The core role of volume calibration models.

[0092] Step 1: Sensor Deployment and Data Acquisition.

[0093] 1. Equipment deployment: A pressure transmitter is installed at the center of the bottom of the storage tank.

[0094] Install a pressure sensor on the top of the storage tank.

[0095] Temperature and humidity sensors were deployed in the tank area.

[0096] All devices are connected to the NB-IoT gateway.

[0097] 2. Key parameter settings: Local atmospheric pressure: =96.0 kPa (96000 Pa); Ambient temperature: =15°C; Average oil temperature: =18°C.

[0098] Storage tank geometric parameters: Horizontal cylindrical tank, diameter D = 3.0 meters, cylinder length L = 15.0 meters, end caps are standard elliptical. Its total volume is calculated to be... Approximately 106.0 cubic meters, maximum liquid level height =D=3.0 meters.

[0099] Step 2: Data preprocessing and transmission.

[0100] At a certain moment, the IoT gateway collects, preprocesses, and uploads the following data packets to the cloud: Compensated tank bottom pressure =116850Pa; Tank top air pressure =96100Pa; oil temperature =18°C.

[0101] Step 3: Cloud-based intelligent inversion calculation (core: irregularly shaped tank processing).

[0102] 1. Volume calibration and shape factor generation for irregularly shaped tanks: The cloud platform first automatically performs virtual sectioning and polynomial fitting based on the input tank geometry (D=3.0m, L=15.0m).

[0103] The height-volume relationship obtained from the fitting is a cubic polynomial (n=3): =106.0×[0.0165×(h / 3.0)^3+0.0358×(h / 3.0)^2+0.9477×(h / 3.0)]; The polynomial has a coefficient of determination R² > 0.999, indicating extremely high accuracy. Therefore, this model can accurately calculate the volume V(h) corresponding to any height h.

[0104] 2. Iterative solution for liquid level inversion: This is key to demonstrating the handling of irregularly shaped cans. The goal is to demonstrate the actual pressure... The actual liquid level H is then determined by solving the equation.

[0105] Given: The density of diesel fuel at 18°C ​​is: ρ(18) = 850 / [1 + 0.0007 × (18 - 20)] ≈ 851.2 kg / m³; Measured pressure difference: =116850-96100=20750Pa; The iterative solution process is as follows: A. Initial guess: Assuming liquid level =2.0 meters.

[0106] B. Query the shape factor: Based on the calibration model, calculate the derivative dV / dh of the volume function V(h) (i.e., the cross-sectional area of ​​the liquid at this moment) when the liquid level is 2.0 meters. The calculated value is dV / dh ≈ 26.18 m². (Tank bottom area) =π×(D / 2)²≈7.07m². Therefore, the shape correction factor is: =26.18 / 7.07≈3.70; It should be noted that when the liquid level reaches 2 meters (approximately 2 / 3 of the 3-meter diameter), the width of the liquid surface in the horizontal tank is much larger than the bottom diameter, resulting in an "equivalent cross-sectional area" that is 3.7 times the bottom area. This is the root cause of the nonlinearity in the hydrostatic pressure-liquid level relationship and is why it is necessary to introduce... The reason for making the correction.

[0107] C. Calculate the theoretical pressure difference: Substitute the guessed value into the formula to calculate the theoretical pressure difference.

[0108] =851.2×9.80665×2.0×3.70≈61780Pa; D. Comparison and Adjustment: The theoretical value (61780Pa) is much larger than the measured value (20750Pa), indicating that the guessed liquid level of 2.0 meters is too high.

[0109] E. Second-order guessing and convergence: The algorithm lowers the guess value. After several iterations, when the guess value... When = 0.85 meters: Found ≈1.85; Theoretical pressure difference calculated: 851.2 × 9.80665 × 0.85 × 1.85 ≈ 13120 Pa; This value is still less than the measured value of 20750 Pa, but it is close. The algorithm will be further fine-tuned, and a dynamic compensation term will be added. ).

[0110] F. Final output: After the algorithm converges, the cloud outputs the precise liquid level value H=1.25 meters.

[0111] 3. Volume calculation: Substituting the liquid level H = 1.25 meters into the calibration polynomial, calculate the current inventory: =106.0×[0.0165×(1.25 / 3.0)^3+0.0358×(1.25 / 3.0)^2+0.9477×(1.25 / 3.0)]≈46.8 cubic meters.

[0112] Step 4: State fusion analysis and anomaly diagnosis.

[0113] The system detected that there were no records of any receiving or dispatching operations in the storage tank in the past 8 hours, but the liquid level slowly dropped from 1.251 meters to 1.250 meters, and the corresponding static pressure difference ΔP dropped by about 15 Pa.

[0114] The algorithm initiates a micro-leakage diagnostic model, fits the pressure data P(t) for that period, and obtains the attenuation trend.

[0115] Although the change was minor, it met the characteristic of "continuous decline when there is no work," so the system marked it as a primary diagnostic state of "extremely low-speed change, attention recommended" and recorded it in the log.

[0116] Step 5: Visualization and Early Warning.

[0117] On the cockpit interface, the horizontal tank is displayed as a 3D model, and its status parameters are as follows: Liquid level: 1.25m (displayed as both percentage and height); Inventory: 46.8 m³; Status: Displayed as "Attention - Static Monitoring". Operators can click to view the detailed diagnostic log for "Very Slow Drop in Liquid Level During Non-Operating Periods".

[0118] System security and optimization: The security mechanisms and model optimization process remain the same as before, ensuring data security and the algorithm's continuous adaptation to the local climate.

[0119] In this invention, it should be noted that, firstly, the formula for calculating the real-time liquid level height H using the improved hydrostatic level inversion model ( In this context, the formula is not a directly calculable arithmetic expression, but rather a physical equation defining the relationship between the liquid level height H and multiple parametric functions. Its principle is based on the fundamental idea of ​​fluid statics (liquid pressure is proportional to liquid column height), but through structural modifications to the denominator and additional terms, three major compensation mechanisms are specifically embedded: The first compensation mechanism is in the denominator. It is for compensating for the geometric deformation of the storage tank, based on the classical static pressure formula ( In this context, it is implicitly assumed that the storage tank is a regular upright column (with a constant cross-sectional area). As a function related to the liquid level height H, it is introduced into the denominator, and its physical meaning is "the ratio of the effective cross-sectional area of ​​the tank at the current liquid level to the reference area". This variable allows the same mathematical model to dynamically describe the real-world variation of the tank's cross-sectional area with height. When the tank is a regular cylinder, ≡1, the formula degenerates into its classical form; when the storage tank is a horizontal tank, spherical tank, or other irregularly shaped tank, As H changes, the nonlinear pressure-level relationship caused by area variation is automatically corrected. This allows the improved hydrostatic level inversion model to be effectively adapted to various tank types.

[0120] The second compensation mechanism is in the supplementary items. It is designed to compensate for the physical deformation of the tank. This is an additional height item independent of pressure calculation. Its principle is based on the fact that the tank body (metal) expands and contracts due to changes in ambient temperature, causing a physical change in the actual height of the tank wall. Even if the pressure measurement is absolutely accurate, the tank's own thermal expansion and contraction will cause the scale to stretch or contract, thus introducing measurement errors. This is equivalent to providing a temperature-based scale calibration for the liquid level results at the algorithm level, which can eliminate systematic errors introduced by the tank material properties and improve environmental adaptability.

[0121] The third compensation mechanism is in the supplementary items. This is a compensation for dynamic interference in the measurement signal. Its principle is to acknowledge the actual measurement signal... In addition to the static liquid column pressure, the result also includes dynamic pressure components caused by fluid sloshing, pump and valve start-up and shutdown vibrations, and random noise. The purpose of this compensation term is to isolate or cancel the influence of these dynamic disturbances from the result, making the final output liquid level value H closer to the true, stable static liquid level. In this way, the stability and anti-interference ability of the monitoring system can be effectively improved, and the reading can remain reliable under operating disturbances.

[0122] Second, for irregularly shaped storage tanks, volume calibration is performed using a height-volume polynomial fitting model. The formula for the height-volume polynomial fitting model is: The summation index k ranges from 0 to n. This formula is not a derivation of a physical law, but rather an engineering modeling method based on mathematical approximation theory. For this formula, firstly, the formula uses... The independent variable is . This is a dimensionless height ratio, mapping the liquid level height of any tank of different sizes to a unified interval of [0,1]. This allows the mathematical model to be independent of the absolute size of a specific tank, becoming a general framework for describing the relationship between relative height and relative volume. Regardless of tank size, its shape characteristics can be described using the same mathematical language (polynomial) within this framework, laying the foundation for the algorithm's universality. Secondly, the formula uses a power function summation form where k ranges from 0 to n. According to Weierstrass's approximation theorem in mathematics, on a closed interval, any continuous function can be uniformly approximated by a polynomial function with arbitrary precision. Thus, no matter how complex (linear, parabolic, S-curve) the actual geometry of the tank (horizontal cylinder, spherical tank, tank with dished heads, etc.) makes its height-volume function (linear, parabolic, S-curve), a set of coefficients can always be found by choosing an appropriate highest order n. This allows the polynomial model V(h) to approximate the true height-volume function with arbitrarily desired engineering accuracy. Finally, a determined set of coefficients... (k=0,1,...,n) uniquely determines a specific polynomial curve. In this model, this set of coefficients... The physical meaning of this is "a mathematical code for the geometry of a specific storage tank". Thus, once the coefficient set of a storage tank is determined through calibration (such as initial geometric parameter calculation or fitting of a few measured points), The complete and continuous height-volume relationship of the storage tank is compressed and stored in these few finite coefficients. Subsequently, for any liquid level h, only a simple polynomial evaluation operation is needed to instantly obtain the high-precision volume V(h), saving the need for massive table lookup storage or complex real-time geometric integration calculations.

[0123] Traditional methods rely on manual climbing, segmented immersion measurements, or complex on-site geometric mapping, which are labor-intensive, dangerous, and require specialized personnel. This invention, through its mathematical model, transforms the calibration process into "input geometric parameters → cloud-based automatic coefficient fitting." The software process is streamlined. On-site personnel do not need complex calibration skills to complete the calibration of any tank type in a short time, effectively reducing the operational threshold, safety risks, and time costs. Meanwhile, traditional mechanical instruments or lookup-based systems often simplify the nonlinear height-volume curve into several broken lines, resulting in jump errors at the segmentation points. The V(h) provided by this model is a value that is accurate across the entire range. An internally continuous, smooth, and infinitely differentiable function can provide an accurate volume value at any height point, fundamentally eliminating the systematic errors caused by piecewise approximations and improving measurement accuracy and the fairness of trade transactions.

[0124] Third, regarding the exponential decay law In this context, the formula describes the dynamic process of pressure release due to leakage in a closed or semi-closed fluid system. Under ideal conditions (small leakage, constant temperature), the pressure change within the system over time follows an exponential decay law. This represents the system pressure at the initial moment (t=0). It is the decay constant, the core quantitative indicator of the entire exponential decay law formula. Its physical meaning is the rate at which pressure decays per unit time. A larger value indicates a faster pressure drop, suggesting a potentially higher rate of leakage. This transforms the vague engineering suspicion of "whether a leak exists" into a precisely calculable mathematical problem: "Can the measured P(t) data be well fitted by an exponential function, and does the fitted λ value exceed a critical value set based on a safety threshold?" This transformation allows leak diagnosis to shift from experience-based, qualitative manual judgment (such as "the pressure seems to be dropping a bit too quickly") to model-based, quantitative, automated analysis.

[0125] In existing technologies, traditional threshold alarms (such as "pressure below a certain value alarm") are static, reactive alarm methods, typically triggered only when leaks have caused significant consequences. This invention, however, identifies the pressure decay trend (λ>0), enabling it to detect system anomalies early in the leak's occurrence, before the absolute pressure value reaches the danger threshold. This significantly advances the warning time, buying valuable time for response and transforming it from "post-event alarm" to "pre-event warning." Furthermore, the decay constant λ is a continuous quantitative indicator. Based on the value of λ, the system can classify the severity of potential risks (e.g., a smaller λ indicates a minor suspected leak, while a larger λ indicates a serious leak warning). This transforms the warning information from a simple "yes / no alarm" to one that includes information on the degree of risk, supporting maintenance personnel in making more precise response decisions (such as scheduling inspections or emergency shutdowns), achieving a leap from alarm to decision support.

[0126] According to a second aspect of the present invention, a low-cost oil depot status monitoring system based on the Internet of Things is also provided for implementing the method described in any of the technical solutions of the first aspect of the present invention, comprising: The sensing and acquisition layer includes several pressure sensors, air pressure sensors, temperature and humidity sensors deployed on the tank body and tank area, as well as an IoT gateway for aggregating sensor data. The network transmission layer employs a narrowband IoT communication module to securely transmit data from the sensing and acquisition layer to the cloud. The cloud-based intelligent processing layer includes: The data access and storage module is used to receive and store monitoring data from the network transport layer; The liquid level inversion and volume calibration calculation module is used to perform real-time liquid level height and volume data calculation; The multi-source data fusion analysis and anomaly diagnosis module is used to perform state fusion analysis and anomaly diagnosis based on real-time liquid level height, volume data and multi-source environmental data. The visualization and early warning service module is used to trigger tiered early warnings based on the results of analysis and anomaly diagnosis, according to preset rules. The application presentation layer provides users with a visual interface to display monitoring results and receive early warning information.

[0127] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An oil depot state low-cost monitoring method based on Internet of Things, characterized in that, The method comprises the following steps: Step S1: deploying a sensor group at the location of the oil depot tank and collecting original monitoring data; Step S2: preprocessing the collected original monitoring data, and then uploading it to the cloud server through the Internet of Things gateway; Step S3: based on the received data, the cloud server inversely calculates the real-time liquid level height of the tank through intelligent algorithms, and calibrates the volume of the special-shaped tank; Step S4: based on the real-time liquid level height, volume data and multi-source environmental data, state fusion analysis and abnormal diagnosis are performed; Step S5: based on the results of analysis and abnormal diagnosis, a hierarchical early warning is triggered according to the preset rules.

2. The low cost monitoring method of oil depot status based on Internet of Things according to claim 1, characterized in that, The sensor group includes a pressure sensor deployed at the bottom of the tank, an air pressure sensor deployed at the top of the tank, and a temperature and humidity sensor deployed in the tank area environment; The original monitoring data includes static pressure data measured by the pressure sensor, atmospheric pressure data measured by the air pressure sensor, and environmental temperature data measured by the temperature and humidity sensor.

3. The low cost monitoring method of oil depot status based on Internet of Things according to claim 2, characterized in that, The step S2 specifically comprises: Compensate the zero drift of the static pressure data based on the real-time environmental temperature to obtain a compensated pressure value; Perform sliding window filtering and outlier rejection based on statistical criteria on the compensated data; Pack the processed data with corresponding environmental data and device identification information, and transmit it to the cloud server through the narrowband Internet of Things network.

4. The low cost monitoring method of oil depot status based on Internet of Things according to claim 3, characterized in that, In step S3, an improved static pressure liquid level inversion model is used to calculate the real-time liquid level height H, and the calculation formula is: wherein P is a compensated pressure value obtained by compensating for zero-point drift, P is the measured atmospheric pressure at the top of the tank, P is the density of the oil product compensated for the current temperature T, P is the acceleration due to gravity, P is a correction factor for the geometry of the tank related to the current level height, P is a compensation term for thermal expansion and contraction of the tank body, P is a dynamic disturbance compensation term based on a Kalman filter estimate.

5. The low cost monitoring method of oil depot status based on Internet of Things according to claim 4, characterized in that, In step S3, for special-shaped tanks, the volume is calibrated through a height-volume polynomial fitting model, and the expression of the model is: wherein, is the liquid volume corresponding to the liquid level height h, is the total volume of the tank, n is a preset parameter selected according to the complexity of the shape of the tank, , k is the sequence variable of the summation term, is the polynomial coefficient obtained by least square fitting, is the maximum height of the tank, and is the tank geometry correction coefficient for a special-shaped tank According to the derivative relationship of the model Dynamic calculation, wherein, is the reference area of the tank bottom of the special-shaped tank.

6. The low cost monitoring method of oil depot status based on Internet of Things according to claim 5, characterized in that, The step S4 specifically comprises: An abnormal diagnosis model is constructed by fusing pressure time series features, liquid level change rate and environmental sensor readings; When the storage tank is determined to be in a static state, the pressure time series data is analyzed. If the fit conforms to the exponential decay law, then... And attenuation constant If the threshold is exceeded, a suspected leak warning will be triggered.

7. The low cost monitoring method of oil depot status based on Internet of Things according to claim 6, characterized in that, The step S5 specifically comprises: Real-time display of the liquid level, inventory, environmental risk index and device health degree of each tank; A three-level warning mechanism of "attention, warning, and serious" is set, and the corresponding level of warning information is automatically triggered according to the output results of the abnormal diagnosis model.

8. The low cost monitoring method of oil depot status based on Internet of Things according to claim 1, characterized in that, The oil depot state low-cost monitoring method based on the Internet of Things further comprises: Step S6: perform end-to-end encryption on the data transmission link, and perform bidirectional certificate authentication on the Internet of Things gateway device.

9. The low cost monitoring method of oil depot status based on Internet of Things as claimed in claim 1 wherein, The oil depot state low-cost monitoring method based on the Internet of Things further comprises: Step S7: the cloud server regularly re-trains the inversion model and abnormal diagnosis model in the intelligent algorithm using the newly added monitoring data.

10. An Internet of Things based low cost monitoring system for tank farm status as claimed in any one of claims 1 to 9, wherein, The method comprises: The sensing and collecting layer comprises a plurality of pressure sensors, air pressure sensors, temperature and humidity sensors deployed on the tank body and the tank area, and an Internet of Things gateway for collecting sensor data; The network transmission layer adopts a narrowband Internet of Things communication module to safely transmit the data of the sensing and collecting layer to the cloud; The cloud intelligent processing layer comprises: A data access and storage module for receiving and storing monitoring data from the network transmission layer; A liquid level inversion and volume calibration calculation module for calculating real-time liquid level height and volume data; The multi-source data fusion analysis and abnormal diagnosis module is used for state fusion analysis and abnormal diagnosis based on real-time liquid level height, volume data and multi-source environmental data; The visualization and early warning service module is used for triggering hierarchical early warning according to preset rules based on the results of analysis and abnormal diagnosis; The application display layer provides a visual operation interface for users, and is used for displaying monitoring results and receiving early warning information.

Images