Intelligent monitoring method and management system for ship pollutant emission

Abstract

The invention relates to the technical field of hygienic products, and discloses an intelligent monitoring method and management system for ship pollutant discharge, and the method comprises the following steps: S1, collecting cabin equipment operation parameters and pollutant discharge data through a multi-source sensor network; s2, an early warning threshold value is dynamically calculated based on the environment working condition and the load state; s3, evaluating the health state of the equipment and predicting potential faults; s4, judging the pollutant discharge compliance; s5, executing a multi-stage control strategy according to evaluation and judgment results; in the step S1, the equipment operation parameters at least comprise temperature, pressure, vibration frequency and oil quality indexes. According to the invention, by integrating various sensors and intelligent processing algorithms, real-time monitoring and dynamic adjustment of ship pollutant emission are realized, and ship emission compliance is ensured; meanwhile, the system can evaluate the health state of the equipment in real time, predict potential faults in advance and automatically execute multi-stage response measures, so that the safety, the environmental protection property and the operation efficiency of the ship are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of sanitary products, and in particular to an intelligent monitoring method and management system for ship pollutant emissions. Background Art

[0002] With the global emphasis on environmental protection, the control of pollutant emissions from ships, as important means of transportation, has become a growing international concern. Especially with the growing global shipping industry, pollutants emitted by ships have become a major source of marine and atmospheric pollution. Pollutants emitted by ships primarily include sulfur oxides (SOx), nitrogen oxides (NOx), hydrocarbons (HC), and particulate matter (PM). These pollutants not only impact the marine ecosystem but also have profound implications for human health, air quality, and global climate change.

[0003] Currently, monitoring of ship pollutant emissions largely relies on traditional manual inspections or simple emissions testing. This approach often fails to provide real-time, accurate emissions data and, due to limitations in crew operational skills, can easily overlook compliance issues with key emission sources. Furthermore, existing emission control systems often lack real-time monitoring of equipment health, preventing potential faults from being identified and addressed in their early stages, impacting ship operational efficiency and safety.

[0004] Most existing emission monitoring technologies focus on measuring a single pollutant, lacking comprehensive, multi-dimensional data collection and real-time analysis capabilities. This often leaves ships' emissions compliance uncontrolled. Existing equipment health monitoring systems, on the other hand, often focus on monitoring a single parameter, lacking comprehensive assessment of the operating status of multiple devices and the ability to predict failures. Existing technologies for maintaining and fault-detecting marine equipment also suffer from issues such as delayed response and low diagnostic accuracy. This results in a delay in implementing effective countermeasures when equipment failures occur, impacting the safety and economic viability of ships.

[0005] Furthermore, the complex and ever-changing operating environment of ships presents numerous challenges for emissions monitoring and real-time monitoring of equipment operating status. For example, as ships operate under varying sea and weather conditions, these changes in the external environment have a direct impact on pollutant emissions. However, traditional monitoring systems do not effectively account for the impact of these environmental factors on emissions behavior and lack flexible adjustment mechanisms to adapt to changing operating conditions. Summary of the Invention

[0006] The purpose of the present invention is to provide an intelligent monitoring method and management system for ship pollutant emissions, which solves the problems in the existing technology that ship pollutant emission monitoring systems cannot monitor ship emissions and equipment status in real time and accurately, and lack dynamic control and intelligent emergency response mechanisms.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: A method for intelligent monitoring of ship pollutant emissions, comprising the following steps: S1. Collect cabin equipment operating parameters and pollutant emission data through a multi-source sensor network; S2. Dynamically calculate warning thresholds based on environmental conditions and load status; S3. Assess equipment health and predict potential failures; S4. Determine the compliance of pollutant emissions; S5. Execute multi-level control strategies based on the evaluation and judgment results.

[0008] Preferably, in S1: The equipment operating parameters include at least temperature, pressure, vibration frequency, and oil quality index; The pollutant emission data includes at least sulfur oxide concentration, nitrogen oxide concentration, and particulate matter content.

[0009] Preferably, the step of dynamically calculating the warning threshold in S2 includes: Calculate the environmental compensation coefficient according to the ambient temperature and humidity; Calculate the load compensation coefficient based on the ratio of the actual load of the equipment to the rated load; Multiply the baseline threshold by the compensation coefficient to obtain the dynamic warning threshold.

[0010] Preferably, the evaluation of the health status of the device in S3 is achieved by the following formula: Where H(t) is the health index, T is the temperature, μ is the oil quality parameter, and α and β are weight coefficients.

[0011] Preferably, the potential fault prediction in S3 adopts a deep residual network model, and the input features include time domain vibration signals and frequency domain energy distribution.

[0012] Preferably, the pollutant emission compliance determination in S4 adopts a diffusion model: Where V is the exhaust flow rate, Q is the diffusion coefficient, C b is the background concentration.

[0013] Preferably, the multi-level control strategy in S5 includes: First-level response: adjust the air-fuel ratio of the combustion chamber; Secondary response: start auxiliary purification device; Level 3 response: Execute equipment load reduction protection.

[0014] Preferably, the step S1 further includes a data preprocessing step: Use sliding window filtering algorithm to eliminate noise interference; Align multi-source sensor data time series using time domain interpolation.

[0015] Preferably, after executing the control strategy, S5 feeds back the execution effect to S2 for dynamic threshold optimization and adjustment.

[0016] An intelligent management system for ship pollutant emissions, comprising: Sensor array modules are deployed in key locations in the cabin; Data communication module, realizing onboard data transmission and ship-to-shore information interaction; Intelligent processing center, a processing unit equipped with intelligent monitoring methods for ship pollutant emissions; The actuator module includes a parameter adjustment device and a multi-level alarm device.

[0017] In summary, the present invention includes at least one of the following beneficial technical effects: 1. This invention utilizes a multi-sensor array to monitor ship pollutant emissions in real time. Combined with the analysis and control algorithms of an intelligent processing center, it accurately determines whether emissions comply with environmental standards. By automatically adjusting combustion parameters, activating auxiliary purification devices, or implementing equipment load reduction measures, this invention ensures that ships maintain compliant pollutant emissions under all operating conditions. This monitoring mechanism significantly improves ship emissions compliance and reduces negative environmental impacts.

[0018] 2. This invention combines health index calculation with deep learning fault prediction models to conduct real-time assessments of the operating status of ship equipment and predict potential faults based on multi-dimensional data (such as temperature, pressure, and vibration). By dynamically monitoring equipment health, the system can proactively identify potential faults and issue alerts, preventing the impact of equipment failures on ship operations and enabling timely initiation of preventive maintenance measures, significantly improving equipment reliability and service life.

[0019] 3. This invention incorporates a multi-stage response mechanism. When a ship's pollutant emissions exceed standards or equipment malfunctions, it can implement a series of automated adjustments, including air-fuel ratio adjustment, auxiliary purification device activation, and equipment load reduction protection, based on the severity of the problem. This mechanism ensures that ship equipment can react quickly to unexpected issues, avoiding delays caused by manual operation, improving ship safety and emergency response capabilities, and ensuring efficient operation in complex environments.

[0020] 4. Through the exchange of information between the ship's data communication module and shore-based information, this invention can transmit ship operating data, pollutant emissions, and equipment health status to a shore-based monitoring center in real time. Shore-based personnel can remotely monitor the ship's operating status and make appropriate decisions, thus achieving intelligent ship management. This data sharing and remote monitoring mechanism effectively improves ship operation efficiency, provides shipping companies with stronger data support, and optimizes ship operation management. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 Schematic diagram of the method of the present invention. DETAILED DESCRIPTION

[0022] The following is combined with Figure 1 , the present invention is described in further detail.

[0023] The present invention provides a method for intelligently monitoring ship pollutant emissions, comprising the following steps: S1. Collect cabin equipment operating parameters and pollutant emission data through a multi-source sensor network; In this embodiment, the multi-source data collection and preprocessing process achieves accurate acquisition of ship equipment operating status and pollutant emission parameters by building an intelligent perception network and data cleaning mechanism. The specific implementation includes the following technical contents: In terms of sensor network deployment, the perception layer of the present invention adopts a multi-type sensor collaborative monitoring solution. In the key equipment parts of the ship's engine room, it is preferred to install a distributed sensor array, including but not limited to: arranging a contact temperature sensor on the outer wall surface of the main engine cylinder liner to obtain the engine operating temperature parameter T in real time. e An online viscosity sensor is integrated into the oil circulation pipeline to calculate the oil viscosity μ0 by measuring the ratio of fluid shear stress to velocity gradient; a non-dispersive infrared spectrometer is installed in the exhaust pipe to estimate the pollutant concentration C by detecting the light intensity attenuation value of a specific wavelength (such as SOx corresponding to the 7.3μm band). SOx .

[0024] To address the synchronous acquisition of multi-source heterogeneous data, this paper designs a time domain alignment processing algorithm. When sensor data with different sampling rates are input, an interpolation method based on the sinc function is used to achieve timing alignment. Its mathematical expression is: Where, represents the aligned continuous signal, x(t k ) is the original data of the kth sampling point, Δt is the reference sampling interval, This process effectively solves the data phase offset problem caused by the difference in sensor response speed.

[0025] In terms of abnormal data detection, the present invention adopts a method based on the three sigma criterion of statistical distribution combined with sliding window analysis. i , the valid data range determination conditions are: Where μ represents the mean of the parameter in the most recent time window and σ is the standard deviation. The marked abnormal data will trigger the sensor self-check procedure and cross-validate with redundant sensor data.

[0026] Preferably, vibration signals are collected using a triaxial accelerometer set, preferably installed in key vibration transmission paths such as the main engine base, gearbox housing, and propulsion shaft bearing seat. The raw vibration signal is first bandpass filtered, with a cutoff frequency preferably set between 10Hz and 5kHz to preserve the equipment's characteristic frequency components.

[0027] For the detection of fuel sulfur content, the present invention sets an online X-ray fluorescence spectrometer in the fuel supply pipeline to calculate the sulfur content S by measuring the characteristic X-ray intensity of sulfur element. f The detection unit is preferably equipped with an automatic calibration module to perform baseline correction using standard samples during the calibration cycle.

[0028] During the data preprocessing phase, the present invention specifically designs a multi-level cache mechanism. The raw data is first stored in a ring buffer and, after preliminary validity verification, transferred to a preprocessing queue. A parallel pipeline processing architecture is employed to perform the following operations: median filtering of the temperature signal to eliminate pulse interference; dimension normalization of the pressure signal; and temperature-pressure compensation calculation of the emission concentration data. The compensation formula is: Where C comp is the concentration value after compensation, P std 、T std is the pressure and temperature under standard working conditions, P axt 、T axt This compensation process effectively eliminates the impact of environmental parameter fluctuations on gas concentration measurement.

[0029] The sensor network preferably uses a dual-bus communication architecture, with key parameters transmitted via the CAN bus and auxiliary parameters transmitted via industrial Ethernet. Each sensor node has a built-in watchdog circuit that automatically activates local data caching when communication is interrupted to ensure data integrity.

[0030] S2. Dynamically calculate warning thresholds based on environmental conditions and load status; In this embodiment, the dynamic warning threshold calculation step achieves precise control over pollutant emission compliance monitoring through a comprehensive analysis of the ship's equipment operating conditions and environmental status. This implementation involves calculating environmental and load compensation coefficients, and optimizing the warning threshold through a dynamic adjustment mechanism to ensure real-time compliance of equipment and pollutant emission parameters.

[0031] First, the present invention adjusts the warning threshold by calculating an environmental compensation coefficient, taking into account the impact of ambient temperature and humidity on pollutant emissions. In practical applications, changes in ambient temperature and humidity have a significant impact on factors such as combustion efficiency and exhaust gas diffusion. This compensation coefficient automatically adjusts the pollutant emission threshold based on environmental changes, ensuring that the system can monitor pollutant emissions in real time according to actual environmental conditions.

[0032] Next, the calculation of the load compensation coefficient reflects the impact of the ship's equipment load on pollutant emissions. Under different load conditions, the emission characteristics of the equipment will vary, so dynamic adjustments need to be made based on the actual load. The calculation formula for the load compensation coefficient is: Among them, γ is the influence coefficient of the load compensation coefficient, L is the actual load of the equipment, L nom is the rated load of the equipment. This calculation method can dynamically adjust the warning threshold of pollutant emissions according to the load status of the equipment, thereby achieving more accurate emission monitoring.

[0033] By calculating environmental and load compensation coefficients, the present invention can adjust pollutant emission monitoring thresholds in real time based on changes in the external environment and equipment load. This dynamic adjustment mechanism adapts to different operating conditions, ensuring that ships can maintain accurate monitoring of pollutant emissions and respond promptly under various operating conditions.

[0034] To further enhance the adaptability of the warning threshold, the present invention's dynamic warning threshold calculation mechanism also incorporates historical equipment operating data. By analyzing this historical data and combining environmental and load compensation factors, the warning threshold can be intelligently optimized. This optimization process automatically adjusts the warning threshold calculation model based on the equipment's long-term operating characteristics and changing trends in pollutant emissions.

[0035] In practice, the warning threshold calculation cycle is a set fixed interval, typically 5 minutes. During this period, the system recalculates the environmental compensation factor and the load compensation factor and adjusts the pollutant emission warning threshold based on these compensation factors. This mechanism enables accurate and dynamic monitoring of pollutant emissions without affecting normal equipment operation.

[0036] Through the above steps, the dynamic warning threshold calculation method in this embodiment achieves efficient and precise adjustment of pollutant emission monitoring. This method can automatically respond to environmental changes and equipment load fluctuations, thereby improving the accuracy and reliability of the ship pollutant emission compliance monitoring system.

[0037] S3. Assess equipment health and predict potential failures; In this embodiment, the equipment status assessment and fault prediction step uses a combination of health index models and deep learning technology to monitor the operating status of ship equipment in real time and predict potential faults, enabling proactive maintenance intervention and minimizing the impact of equipment failures on ship operations. The specific implementation of this step involves the calculation of health indexes and the design and application of fault prediction models.

[0038] First, the health index (HI) is used to reflect the overall health status of ship equipment. The calculation of the health index is based on the dynamic weighted integration of multiple key parameters during equipment operation, including temperature, pressure, vibration frequency and other parameters. Specifically, the health index H(t) is derived from the integration of multiple physical quantities of the equipment, and its calculation formula is: Among them, T represents the temperature parameter, μ i is the i-th lubricating oil quality parameter, w i is the weight coefficient of this parameter, α and β are the influence coefficients of temperature change and oil quality parameters on the health index, respectively. The health index reflects the overall operating health of the equipment and can effectively determine whether the equipment is within the normal operating range.

[0039] The first item in the health index is the integral of temperature change, which indicates how the device's temperature changes over long-term operation. This integral of temperature change can be used to assess the device's operational stability under different operating conditions. Large temperature fluctuations indicate a high workload or potential failure.

[0040] The second component of the health index is a weighted sum of lubricant quality parameters, reflecting the operating condition of the equipment's lubricant. Lubricant quality is crucial to the equipment's lubrication performance. Factors such as lubricant viscosity, acidity, and metal content all affect equipment wear and failure. By combining these lubricant quality parameters, it is possible to determine whether the equipment is in a normal lubrication state.

[0041] Secondly, the potential fault prediction model is designed through deep learning technology in order to predict the type of failure that may occur in the equipment in advance. To this end, the present invention uses a deep residual network (DRN) to analyze and predict the status of the equipment. The input features include the time domain vibration signal and frequency domain energy distribution of the equipment. Specifically, after the input data is feature extracted, it enters the deep residual network for training and prediction. The structure of the deep residual network is constructed through multiple layers of residual blocks, each layer of which contains a convolution layer, a batch normalization layer, and an activation function layer, so as to effectively learn the complex relationship between the operating mode and the failure mode of the equipment.

[0042] The structure of the deep residual network can be expressed as: in, is the residual output of the input data x, is the transformation of the input data after convolution and activation function, The residual between the original input data and the transformed result. Through residual learning, deep networks can more accurately capture the characteristic changes of equipment under different working conditions, thereby predicting equipment failures.

[0043] Specific fault predictions include a variety of possible fault types, such as bearing wear, piston ring fracture, and gearbox failure. Each fault type is associated with a separate output probability value, indicating the likelihood of the equipment failing within a certain period of time. Using these output probability values, the system can issue timely warnings and provide a basis for subsequent maintenance decisions.

[0044] By combining the aforementioned health index calculation with deep learning-based fault prediction, this embodiment enables real-time assessment of the health status of ship equipment and early warning of potential faults. The system comprehensively monitors the operating status of equipment based on real-time data and uses fault prediction models to identify possible fault types, enabling proactive maintenance measures to mitigate the impact of equipment failures on ship safety and economic efficiency.

[0045] S4. Determine the compliance of pollutant emissions; In this embodiment, the pollutant emission compliance determination step aims to monitor pollutants emitted by ships in real time and determine whether emissions meet preset standards through a comprehensive analysis of environmental factors, equipment status, and emission parameters. This step ensures that pollutant emissions remain within safe and compliant limits by constructing a pollutant diffusion model and adapting it to environmental conditions in real time.

[0046] First, emissions compliance determination relies on a mathematical model based on gaseous diffusion, which accounts for the interaction between a ship's emission source and the surrounding environment. Specifically, a Gaussian plume model is used to describe the diffusion of pollutants in the atmosphere. This model uses input variables such as the location of the pollutant emission source, the intensity of the pollutant emission, ambient wind speed, and the gas diffusion coefficient to calculate the concentration distribution of pollutants at different locations.

[0047] The basic form of the diffusion model used is: Among them, C(x,y,z) represents the pollutant concentration at any point in three-dimensional space, Q is the pollutant emission intensity, u is the wind speed, σ y and σ z where ∠ and ∠ are the horizontal and vertical diffusion coefficients, respectively, and H is the height of the emission source. The model calculates the diffusion and concentration distribution of pollutants by considering the emission source's location, wind speed, and meteorological conditions (such as temperature and humidity). The pollutant concentration values ​​output by the model are compared with pre-set emission compliance standards to determine whether current emissions are in compliance.

[0048] During the compliance determination process, the system will monitor the concentration of the exhaust gas in real time and compare it with the pollutant concentration calculated based on the above diffusion model. If the real-time monitored pollutant concentration exceeds the maximum safe concentration C calculated by the diffusion model, max , it is considered that the ship’s emissions do not meet environmental protection requirements and the alarm mechanism is triggered.

[0049] The specific compliance determination also needs to consider the background pollutant concentration C b , that is, the background concentration of pollutants caused by external environmental factors (such as surrounding ships, industrial areas, etc.). Background concentration C b This will be taken into consideration during the judgment process to ensure that the model results are closer to the actual situation. Based on this, the judgment formula for pollutant concentration is revised to: Where V is the exhaust flow rate, and k1 and k2 are constant parameters related to wind speed, air pressure, and other environmental conditions. This formula dynamically adjusts the safe emission threshold of pollutants, ensuring the accuracy and adaptability of compliance determinations as environmental conditions change.

[0050] In actual use, environmental data (such as wind speed, temperature, and humidity) and emissions data are collected and input into the system through real-time sensors. Based on this real-time data, the system calculates environmental compensation coefficients and pollutant diffusion coefficients and dynamically adjusts compliance thresholds. This dynamic adjustment mechanism ensures the real-time and accurate determination of emissions compliance.

[0051] In addition, this embodiment incorporates a data correction mechanism to account for potential measurement errors and transient fluctuations caused by environmental changes. If the system detects abnormal fluctuations in pollutant concentration (such as transient errors caused by sudden changes in wind speed or short-term fluctuations in equipment load), it filters the data using a sliding window smoothing algorithm. This processing algorithm performs a weighted average of pollutant concentration data over a period of time to filter out noise and improve the stability of monitoring results.

[0052] In summary, the pollutant emission compliance determination in this embodiment, through the combination of a Gaussian plume model and real-time environmental data, enables precise monitoring of ship pollutant emissions and dynamically adjusts emission compliance thresholds during ship operation, ensuring that ship emissions meet environmental standards. This mechanism effectively improves the system's ability to monitor emission compliance and, during actual operations, provides timely feedback on potential emissions exceeding standards, assisting ship operators in taking appropriate measures.

[0053] S5. Execute multi-level control strategies based on the evaluation and determination results; In this embodiment, the multi-level control strategy execution step uses real-time analysis of the ship's equipment operating status, emissions compliance, and potential faults, combined with pre-defined control strategies to implement multi-level responses, thereby taking appropriate emergency measures for varying degrees of abnormality. This step is intended to ensure the safety and environmental compliance of ship equipment operations, while optimizing equipment operating status through automated adjustments.

[0054] First, the present invention establishes a multi-level control response mechanism through real-time monitoring of pollutant emissions and equipment health data. When the system detects potential risks of exceeding pollutant emissions standards or equipment failure, it sequentially implements different levels of response measures based on pre-set warning levels. These measures include, but are not limited to, adjusting the combustion chamber air-fuel ratio, activating auxiliary purification devices, and, when necessary, implementing equipment load reduction protection.

[0055] If pollutant emission compliance fails to meet requirements, the primary response mechanism is triggered. This mechanism preferentially adjusts the air-fuel ratio of the ship's main engine to optimize combustion efficiency and reduce the generation of harmful substances. The basic principle of air-fuel ratio adjustment is to increase the amount of air to improve combustion efficiency, thereby reducing pollutant emissions.

[0056] If the excessive emissions are not resolved through the primary response, the system will trigger a secondary response. During this phase, the system preferably activates auxiliary purification devices, such as seawater scrubbers, for further purification. Scrubbers remove harmful substances such as sulfur oxides from the exhaust gas by reacting seawater with pollutants. The seawater scrubber's workflow involves controlling the circulation pump flow rate and the amount of lye injected. These flow rates and lye volume are preferably dynamically adjusted based on pollutant concentration to maximize purification efficiency.

[0057] If pollutant emissions still fail to meet compliance standards and the equipment health status indicates a potential failure risk, the system will initiate a three-level response mechanism, namely, equipment load reduction protection. Load reduction protection reduces equipment load by limiting the main engine speed, reducing the risk of excessive emissions and equipment failure caused by excessive load. At this time, the system adjusts the engine power output to keep the ship operating within a safe range. The mathematical model of load reduction control is: P load =f(L max ,T current ,P target ); Among them, P load is the current load, L max is the maximum load of the equipment, T current is the current temperature of the device, P target The target load is the load. Through this control model, the system can automatically adjust the load according to the working status of the equipment, avoiding the equipment from operating under extreme conditions and reducing the probability of failure.

[0058] The core of the multi-level control strategy lies in the hierarchical and dynamic adjustment of responses. The system monitors the equipment's operating status, emissions data, and health index in real time to select the most appropriate response measures under different operating conditions. Each level of response is adjusted based on the effectiveness of the previous level to ensure optimal equipment operation and minimize environmental impact.

[0059] In addition, the present invention also designs a feedback mechanism for the control strategy. During the multi-level response execution process, the system will collect execution effect data in real time and optimize the control strategy through the feedback mechanism. Specifically, when the first and second level responses are activated, the system will continuously monitor parameters such as emission values, equipment temperature and pressure, and feed this data back to the control system to assess whether further strengthening or adjustment of control measures is needed. The mathematical model of feedback optimization is: Where, ΔC feedback is the feedback adjustment amount, C measured is the current measured value, C target is the target value, α iThis feedback mechanism ensures the accuracy of each control strategy and continuously optimizes the control logic during long-term system operation.

[0060] In summary, the multi-level control strategy in this embodiment uses a comprehensive analysis of pollutant emissions, equipment health, and workload to dynamically adjust and respond to different operating conditions in a graded manner, ensuring optimal equipment operation and minimizing environmental impact. The flexibility and adaptability of this control strategy enable the system to efficiently handle a variety of complex operating conditions, improving the ship's environmental compliance and equipment operational safety.

[0061] An intelligent management system for ship pollutant emissions, comprising: Sensor array modules are deployed in key locations in the cabin; The sensor array modules of this system are placed in key equipment locations in the ship's engine room, such as the main engine, auxiliary engine, boiler, exhaust pipe and other important locations. The sensor array modules include but are not limited to the following types of sensors: Temperature sensors: These are used to monitor the temperature of ship equipment in real time, particularly engines and other high-temperature components. The sensors are positioned to ensure coverage of all heat sources and critical equipment, reflecting the equipment's operating status.

[0062] Pressure sensors: These monitor pressure changes in combustion chambers, oil lines, and gas pipelines. By monitoring pressure fluctuations, the system can determine if the equipment is experiencing abnormal loads or a decrease in operating efficiency.

[0063] Pollutant sensors, such as non-dispersive infrared (NDIR) sensors, are used to detect sulfur oxides (SOx), nitrogen oxides (NOx), and particulate matter (PM) concentrations in exhaust gases. The selection and placement of sensors ensure accurate monitoring of pollutant levels emitted by ships.

[0064] Vibration sensor: used to monitor the vibration condition of the equipment in real time and promptly detect abnormal vibration of the equipment, such as bearing damage, gear wear and other potential faults.

[0065] The layout and number of sensor array modules can be flexibly adjusted according to the specific type of ship and the operating environment to ensure comprehensive coverage of equipment and emission sources and real-time collection of relevant data.

[0066] Data communication module, realizing onboard data transmission and ship-to-shore information interaction; The data communication module is responsible for realizing data transmission between each sensor node in the ship and the intelligent processing center, and ensuring real-time information exchange between the ship and the shore. The communication module mainly includes the following components: Onboard LAN: This uses efficient and stable communication protocols (such as CAN, Modbus, and Ethernet) to transmit sensor data to the intelligent processing center. The onboard LAN supports multi-device data transmission and processing and provides redundant channels to ensure high data transmission reliability.

[0067] Ship-to-shore communication interface: Via satellite communications or 4G / 5G networks, the system can transmit real-time ship data to shore, facilitating remote monitoring and data analysis by shipping companies and relevant regulatory authorities. The data communication module also supports data encryption and authentication to ensure data security.

[0068] Wireless transmission module: In specific application scenarios, a wireless data transmission module can be optionally installed to achieve flexible communication between devices and reduce wiring complexity.

[0069] Through the data communication module, the system can realize real-time information transmission and remote management, ensuring the efficiency of ship pollutant emission monitoring and equipment maintenance.

[0070] Intelligent processing center, a processing unit equipped with intelligent monitoring methods for ship pollutant emissions; The Intelligent Processing Center (IPC) is the core of the intelligent management system for ship pollutant emissions. It is responsible for receiving data from the sensor array modules, processing and analyzing the data, and responding based on the analysis results. The IPC's main functional modules include: The Data Receiving and Preprocessing Unit performs preliminary processing on sensor data, including data filtering, denoising, anomaly detection, and time series alignment. This unit ensures the accuracy and availability of input data.

[0071] Data analysis and decision-making unit: Based on real-time monitoring of equipment status, pollutant emission concentration and other data, algorithm models (such as health index calculation, fault prediction model, emission compliance determination model, etc.) are used to judge the ship's operating status and environmental compliance.

[0072] Early warning and alarm unit: Based on the data analysis results, when the system detects equipment failure or pollutant emissions exceeding the standard, the early warning and alarm unit will generate corresponding alarm information and issue an alarm to the crew or shore-based system through the onboard control system or data communication module.

[0073] The functional units of the intelligent processing center integrate advanced computing technology and intelligent algorithms to efficiently evaluate the ship's operating conditions in real time, provide early warnings before potential problems occur, and help crew members take preventive measures.

[0074] Actuator module, including parameter adjustment device and multi-level alarm device; The actuator module is responsible for executing various regulatory operations according to the instructions of the intelligent processing center and ensuring the normal operation of ship equipment and compliance with pollutant emissions. The actuator module includes the following two functional units: Parameter adjustment devices: These devices adjust the operating parameters of marine equipment based on real-time monitoring data. For example, they can adjust the main engine's air-fuel ratio to reduce pollutant emissions or tweak the fuel injection system to optimize combustion. These devices can be connected to equipment via an automatic control system to adjust operating parameters in real time.

[0075] Multi-level alarms: These trigger alarms in the event of excessive pollutant emissions or equipment failure. These preferably include audible and visual alarms, vibration alarms, and a remote warning system to ensure crew members can respond promptly and prevent the impact of substandard pollutant emissions or equipment failures.

[0076] The actuator module is closely connected to the intelligent processing center. Through two-way data interaction, it ensures that the system can execute corresponding control instructions based on the processing results, maintaining the efficient operation and environmental compliance of the ship.

[0077] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for intelligent monitoring of ship pollutant emissions, characterized in that: The following steps are involved: S1. Collect cabin equipment operating parameters and pollutant emission data through a multi-source sensor network; S2. Dynamically calculate warning thresholds based on environmental conditions and load status; S3. Assess equipment health and predict potential failures; S4. Determine the compliance of pollutant emissions; S5. Execute multi-level control strategies based on the evaluation and judgment results.

2. The intelligent monitoring method for ship pollutant emissions according to claim 1, characterized in that: In S1: The equipment operating parameters include at least temperature, pressure, vibration frequency, and oil quality index; The pollutant emission data includes at least sulfur oxide concentration, nitrogen oxide concentration, and particulate matter content.

3. The intelligent monitoring method for ship pollutant emissions according to claim 1 is characterized in that: The step of dynamically calculating the warning threshold in S2 includes: Calculate the environmental compensation coefficient according to the ambient temperature and humidity; Calculate the load compensation coefficient based on the ratio of the actual load of the equipment to the rated load; Multiply the baseline threshold by the compensation coefficient to obtain the dynamic warning threshold.

4. The intelligent monitoring method for ship pollutant emissions according to claim 1, characterized in that: The evaluation of the equipment health status in S3 is achieved by the following formula: Where H(t) is the health index, T is the temperature, μ is the oil quality parameter, and α and β are weight coefficients.

5. The intelligent monitoring method for ship pollutant emissions according to claim 1 is characterized in that: The potential fault prediction in S3 adopts a deep residual network model, and the input features include time domain vibration signals and frequency domain energy distribution.

6. The intelligent monitoring method for ship pollutant emissions according to claim 1 is characterized in that: The pollutant emission compliance determination in S4 adopts the diffusion model: Where V is the exhaust flow rate, Q is the diffusion coefficient, C b is the background concentration.

7. The intelligent monitoring method for ship pollutant emissions according to claim 1, characterized in that: The multi-level control strategy in S5 includes: First-level response: adjust the air-fuel ratio of the combustion chamber; Secondary response: start auxiliary purification device; Level 3 response: Execute equipment load reduction protection.

8. The intelligent monitoring method for ship pollutant emissions according to claim 1 is characterized in that: The S1 also includes the following data preprocessing steps: Use sliding window filtering algorithm to eliminate noise interference; Align multi-source sensor data time series using time domain interpolation.

9. The intelligent monitoring method for ship pollutant emissions according to claim 1, characterized in that: After S5 executes the control strategy, the execution effect is fed back to S2 for dynamic threshold optimization and adjustment.

10. An intelligent management system for ship pollutant emissions, used in accordance with the intelligent monitoring method for ship pollutant emissions according to any one of claims 1 to 9, characterized in that: include: Sensor array modules are deployed in key locations in the cabin; Data communication module, realizing onboard data transmission and ship-to-shore information interaction; An intelligent processing center, configured with a processing unit of the method according to any one of claims 1 to 9; The actuator module includes a parameter adjustment device and a multi-level alarm device.

Images