A ship exhaust gas treatment system full life cycle predictive maintenance method and system

Abstract

The application discloses a kind of ship exhaust treatment system full life cycle predictive maintenance method and system, specifically relates to ship exhaust treatment technical field, by introducing thermal inertia compensation and time reconstruction mechanism, the exhaust temperature and emission data mapped to the equivalent reaction temperature evolution quantity capable of representing the real heating state inside catalytic converter obtained by limited sensor, combined with transient thermal shock intensity factor and thermal working condition segmented state vector, effective differentiation to strong transient condition and quasi-steady state condition is realized, the collaborative discrimination system of transient function deviation index and long-term degradation cumulative index is constructed, so as to be able to accurately distinguish short-time emission fluctuation caused by control lag and thermal response mismatch and irreversible material degradation caused by repeated thermal shock accumulation under complex navigation condition.

Description

Technical Field

[0001] This invention relates to the field of marine exhaust gas treatment technology, and more specifically, to a predictive maintenance method and system for the entire life cycle of a marine exhaust gas treatment system. Background Technology

[0002] During actual navigation, ships often operate under complex sea states and unsteady maneuvering conditions. Affected by factors such as surges, rapid acceleration, and frequent speed changes, the main engine load, exhaust gas flow rate, and exhaust temperature exhibit significant and strong transient changes, placing the ship's exhaust gas treatment system under constant thermal shock and rapid operating condition switching. Especially within core units such as SCR and DPF, the catalytic reaction and particulate filtration processes are highly sensitive to temperature field distribution. However, existing monitoring systems typically rely on a limited number of temperature and emission sensors, acquiring mostly local point information, which is insufficient to accurately reflect the spatial temperature gradient and thermal inertia effects formed within the catalytic bed under transient conditions. When operating conditions change rapidly, the sensor response delay and heat conduction lag combine, easily causing a misalignment between monitored parameters and the actual physical state, thus causing the degradation model input to deviate from the actual material and structural heating state. Furthermore, this misalignment is amplified under strong transient conditions, manifesting as abnormal fluctuations in NOx emission indicators within a short period. However, these fluctuations often stem from insufficient injection control response or an unstable catalytic reaction, and do not necessarily correspond to structural degradation of the catalytic material. However, existing maintenance methods are mostly based on empirical judgments of emission indicators or pressure difference changes. They are difficult to effectively distinguish between functional fluctuations caused by physical transients and irreversible material damage caused by long-term thermal shock accumulation in scenarios with frequent sudden changes in operating conditions. This can easily lead to misjudgment or delayed maintenance, which restricts the engineering applicability and reliability of predictive maintenance of ship exhaust gas treatment systems in real navigation environments. Summary of the Invention

[0003] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a predictive maintenance method and system for the entire lifecycle of a ship exhaust gas treatment system, in order to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: A predictive maintenance method for the entire life cycle of a ship exhaust gas treatment system includes the following steps: collecting operational data of the ship exhaust gas treatment system under different navigation stages, and calculating the thermal inertia parameters and thermal diffusion time constant of the catalyst by combining the catalyst structural dimensions, material heat capacity and thermal conductivity coefficient, to obtain the thermal response parameter set of the exhaust gas treatment unit; Based on the thermal response parameter set, thermal inertia compensation and time reconstruction are performed on the collected exhaust temperature sequence to calculate the equivalent reaction temperature evolution, and the exhaust temperature change rate and load change rate are introduced to construct the transient thermal shock intensity factor. Based on the transient thermal shock intensity factor and the evolution of the equivalent reaction temperature, the thermal state evolution characteristics of the exhaust gas treatment system during operation are extracted, and a segmented thermal condition state vector including rapid thermal transition state and quasi-steady-state reaction state is constructed. Based on the segmented state vector of thermal conditions, the transient function deviation index and the long-term degradation accumulation index are calculated by combining the nitrogen oxide conversion efficiency, injection response deviation and pressure difference change trend. Based on the synergistic relationship between the transient functional deviation index and the long-term degradation cumulative index, a health evolution discrimination model for the exhaust gas treatment system is constructed. When an emission anomaly is detected as being caused by transient thermal mismatch, maintenance is triggered to suppress it. When the long-term degradation cumulative index shows a continuous increase and approaches the material tolerance limit, it is determined that the system has an accelerated life consumption trend, and predictive maintenance early warning information is generated.

[0005] In a preferred embodiment, the process of collecting operational data of the ship's exhaust gas treatment system at different navigation stages, and calculating the catalyst's thermal inertia parameters and thermal diffusion time constant by combining the catalyst's structural dimensions, material heat capacity, and thermal conductivity coefficient, to obtain the thermal response parameter set of the exhaust gas treatment unit is as follows: In the ship exhaust gas treatment system, high-temperature corrosion-resistant temperature sensors, differential pressure sensors and nitrogen oxide concentration sensors are arranged at the exhaust gas inlet and outlet, and the main engine load signal, exhaust flow signal and injection control command are obtained in the main engine control system. The temperature sensor is used to collect the inlet temperature of the exhaust gas treatment system. With outlet temperature The differential pressure sensor is used to collect the pressure difference before and after the particle trap. The nitrogen oxide concentration sensor is used to collect nitrogen oxide concentration values ​​before and after exhaust gas treatment; The collected host load, exhaust flow, temperature data, differential pressure data, nitrogen oxide concentration data and injection control commands are unified with timestamps and time-aligned according to the preset sampling period to form the original operating data sequence of the exhaust gas treatment system. By combining the structural design parameters of the exhaust gas treatment system, the density and volume of different materials in the catalyst are obtained, and the heat capacity parameters of the catalyst are calculated by combining the specific heat capacity of the catalyst materials. Based on the axial length of the catalyst and the thermal conductivity of the material, the characteristic time of heat diffusion along the axial direction of the catalyst is calculated, and the thermal diffusion time constant of the catalyst bed is obtained. The heat capacity parameter and the thermal diffusion time constant are combined to construct a set of thermal response parameters for the exhaust gas treatment unit.

[0006] In a preferred embodiment, the process of performing thermal inertia compensation and time reconstruction on the collected exhaust temperature sequence based on the thermal response parameter set, calculating the equivalent reaction temperature evolution, and introducing the exhaust temperature change rate and load change rate to construct the transient thermal shock intensity factor is as follows: From the set of thermal response parameters, the thermal capacity parameters and thermal diffusion time constant of the catalyst are obtained and used as physical constraint parameters to describe the thermal inertial characteristics of the exhaust gas treatment unit. The collected inlet temperature of the exhaust gas treatment system With outlet temperature The exhaust temperature time series data is formed by arranging the data according to a unified timestamp; Based on the first-order thermal inertia model, thermal inertia compensation is applied to the exhaust temperature time series, and its equivalent reaction temperature is obtained. The evolutionary relationship can be expressed as: ,in The thermal diffusion time constant; Under discrete-time conditions, the equivalent reaction temperature The evolution relationship was numerically reconstructed to obtain the equivalent reaction temperature evolution sequence; Combined with exhaust temperature change rate The calculation method is as follows: ,in The time interval between adjacent sampling times; Synchronous calculation of host load change rate The calculation method is as follows: ,in The load level of the host at time t; After normalizing the exhaust temperature change rate and the load change rate, a weighted fusion was performed to construct the transient thermal shock intensity factor.

[0007] In a preferred embodiment, the process of extracting the thermal state evolution characteristics of the exhaust gas treatment system during operation based on the transient thermal shock intensity factor and the equivalent reaction temperature evolution, and constructing a segmented thermal condition state vector including rapid thermal transition states and quasi-steady-state reaction states, is as follows: Based on the equivalent reaction temperature evolution and the transient thermal shock intensity factor, the thermal state changes of the exhaust gas treatment system during continuous operation are jointly analyzed. A sliding window is used to divide the equivalent reaction temperature sequence on a continuous time axis, dividing the process into several adjacent time windows. Where k represents the time window number; Within each time window, the average change in equivalent reaction temperature and the mean value of transient thermal shock intensity factor are calculated respectively. The stability index of the rate of change of the equivalent reaction temperature within the time window is calculated and expressed by the standard deviation of the first difference of the equivalent reaction temperature; The average variation of the equivalent reaction temperature, the mean of the transient thermal shock intensity factor, and the standard deviation of the first difference of the equivalent reaction temperature are combined as thermal state evolution characteristic quantities to construct the thermal condition state vector corresponding to the time window. Based on the thermal condition state vector and the preset thermal shock judgment threshold, the thermal condition type of each time window is classified and judged; the thermal shock judgment threshold includes the equivalent reaction temperature change amplitude threshold, the transient thermal shock intensity factor threshold, and the temperature change rate stability standard deviation threshold. When the average change of equivalent reaction temperature is greater than or equal to the threshold of equivalent reaction temperature change, the mean of transient thermal shock intensity factor is greater than or equal to the threshold of transient thermal shock intensity factor, and the standard deviation of the first difference of equivalent reaction temperature is greater than or equal to the standard deviation threshold of temperature change rate stability, the time window is determined to be in a rapid thermal transition state. When the average change of equivalent reaction temperature is less than the threshold of equivalent reaction temperature change, the mean of transient thermal shock intensity factor is less than the threshold of transient thermal shock intensity factor, and the standard deviation of the first difference of equivalent reaction temperature is less than the standard deviation threshold of temperature change rate stability, the time window is determined to be in a quasi-steady-state reaction state. The thermal condition type and thermal condition state vector corresponding to each time window are combined in chronological order to form a segmented state vector sequence of thermal conditions for the exhaust gas treatment system.

[0008] In a preferred embodiment, the process of calculating the transient performance deviation index and the long-term degradation accumulation index based on the segmented state vector of thermal conditions, combined with the nitrogen oxide conversion efficiency, injection response deviation, and pressure difference change trend, is as follows: From the segmented state vector sequence of thermal conditions, obtain the thermal condition type identifier corresponding to each time window to distinguish between rapid thermal transition state and quasi-steady-state reaction state. Within each time window, the nitrogen oxide conversion efficiency is calculated based on the inlet and outlet nitrogen oxide concentration data; Simultaneously acquire the injection control command and actual injection response within the time window, and calculate the injection response deviation; Within the time window determined to be a rapid thermal transition state, a transient functional deviation index is constructed based on the nitrogen oxide conversion efficiency and injection response deviation. Within the time window determined to be a quasi-steady-state reaction state, the difference between the average pressure difference of adjacent time windows is used to represent the trend of pressure difference changes before and after the particle trap. By combining the aforementioned pressure difference change trend with the long-term evolution characteristics of nitrogen oxide conversion efficiency, a long-term degradation cumulative index is constructed. The transient functional deviation index and the long-term degradation cumulative index are associated with the corresponding thermal condition segmented state vectors to form a functional deviation-degradation evolution feature set of the exhaust gas treatment system under different operating conditions.

[0009] In a preferred embodiment, a health evolution discrimination model for the exhaust gas treatment system is constructed based on the synergistic relationship between the transient functional deviation index and the long-term degradation accumulation index. When an emission anomaly is detected as being caused by transient thermal mismatch, maintenance is inhibited. When the long-term degradation accumulation index shows a continuous increase and approaches the material tolerance limit, it is determined that the system has an accelerated lifespan consumption trend. The process of generating predictive maintenance early warning information is as follows: Based on the transient functional deviation index and the long-term degradation cumulative index, a dual-index health evolution characteristic space of the exhaust gas treatment system during operation is constructed. In the dual-index health evolution feature space, a health status discrimination function is introduced; Based on historical operating data and material tolerance characteristics, a transient function deviation threshold range is preset in the dual-index feature space. With long-term degradation risk threshold ; When transient functional deviation from the index is detected within a consecutive number of time windows The levels continued to rise and all exceeded the maximum acceptable range for short-term emission anomalies. However, the long-term cumulative degradation index Less than or equal to the preset long-term degradation risk threshold At that time, it was determined that the current emission anomaly was caused by transient thermal mismatch due to thermal inertia and control response lag; Under the above-mentioned conditions, the health evolution discrimination model enters the maintenance suppression state and suppresses the maintenance trigger signal. When the long-term degradation cumulative index is detected within multiple consecutive quasi-steady-state reaction time windows... It exhibits a monotonically increasing trend, and the value is greater than the preset long-term degradation risk threshold. At that time, it was determined that the exhaust gas treatment system showed a trend of accelerated lifespan consumption; After determining that there is a trend of accelerated lifespan depletion, the health evolution discriminant model generates predictive maintenance early warning information.

[0010] In a preferred embodiment, a predictive maintenance system for the entire life cycle of a ship exhaust gas treatment system includes an operation perception modeling module, a thermal response reconstruction module, a thermal condition segmentation module, a degradation dual-index assessment module, and a health judgment decision module. The perception modeling module is used to collect operational data of the ship's exhaust gas treatment system at different navigation stages. Combined with the catalyst structure size, material heat capacity and thermal conductivity, the thermal inertia parameters and thermal diffusion time constant of the catalyst are calculated to obtain the thermal response parameter set of the exhaust gas treatment unit. The thermal response reconstruction module is used to perform thermal inertia compensation and time reconstruction on the collected exhaust temperature sequence according to the thermal response parameter set, calculate the equivalent reaction temperature evolution, and introduce the exhaust temperature change rate and load change rate to construct the transient thermal shock intensity factor. The thermal condition segmentation module is used to extract the thermal state evolution characteristics of the exhaust gas treatment system during operation based on the transient thermal shock intensity factor and the equivalent reaction temperature evolution, and to construct a thermal condition segmentation state vector that includes rapid thermal transition state and quasi-steady-state reaction state. The degradation dual-index assessment module is used to calculate the transient functional deviation index and the long-term degradation cumulative index based on the segmented state vector of thermal conditions, combined with the nitrogen oxide conversion efficiency, injection response deviation and pressure difference change trend. The health assessment and decision-making module is used to construct a health evolution assessment model for the exhaust gas treatment system based on the synergistic relationship between the transient functional deviation index and the long-term degradation cumulative index. When an emission anomaly is detected as being caused by transient thermal mismatch, maintenance is triggered to suppress it. When the long-term degradation cumulative index shows a continuous increase and approaches the material tolerance limit, it is determined that the system has an accelerated life consumption trend, and predictive maintenance early warning information is generated.

[0011] The technical effects and advantages of this invention are as follows: 1. This invention introduces a thermal inertia compensation and time reconstruction mechanism to map the exhaust temperature and emission data obtained by limited sensors into an equivalent reaction temperature evolution that can characterize the actual heating state inside the catalyst. Combined with the transient thermal shock intensity factor and the segmented state vector of thermal conditions, it achieves effective differentiation between strong transient conditions and quasi-steady-state conditions. It constructs a collaborative discrimination system of transient functional deviation index and long-term degradation accumulation index, thereby accurately distinguishing between short-term emission fluctuations caused by control lag and thermal response mismatch and irreversible material degradation caused by repeated thermal shock accumulation under complex navigation conditions.

[0012] 2. This invention avoids the risk of misjudgment caused by relying solely on experience-based judgments based on a single emission or pressure difference index, significantly reduces the probability of triggering unnecessary maintenance, and can provide early warnings before the accelerated consumption of catalytic materials and filter structures, achieving truly predictive maintenance throughout the entire life cycle. This improves the operational reliability, maintenance decision accuracy, and engineering applicability of ship exhaust gas treatment systems in real navigation environments. Attached Figure Description

[0013] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings; Figure 1 This is a flowchart of the method in Embodiment 1 of the present invention; Figure 2 This is a flowchart of the system in Embodiment 2 of the present invention. Detailed Implementation

[0014] 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 only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0015] Example 1: Figure 1 This invention presents a predictive maintenance method for the entire lifecycle of a ship's exhaust gas treatment system, comprising the following steps: collecting operational data of the ship's exhaust gas treatment system under different navigation stages, including main engine load, exhaust flow rate, inlet and outlet temperatures of the exhaust gas treatment system, particulate filter pressure difference, nitrogen oxide concentration, and injection control commands; and combining the catalyst structure dimensions, material heat capacity, and thermal conductivity coefficient to calculate the catalyst thermal inertia parameters and thermal diffusion time constant, thereby obtaining the thermal response parameter set of the exhaust gas treatment unit; Based on the thermal response parameter set, thermal inertia compensation and time reconstruction are performed on the collected exhaust temperature sequence to calculate the equivalent reaction temperature evolution that reflects the true internal thermal state. The exhaust temperature change rate and load change rate are introduced to construct a transient thermal shock intensity factor to characterize the degree of thermal shock under frequent acceleration and deceleration conditions. Based on the transient thermal shock intensity factor and the evolution of the equivalent reaction temperature, the thermal state evolution characteristics of the exhaust gas treatment system during operation are extracted, and a segmented thermal operating condition state vector including rapid thermal transition state and quasi-steady-state reaction state is constructed to distinguish between the strong transient operating condition stage and the continuous stable operating condition stage. Based on the segmented state vector of thermal conditions, and combined with the nitrogen oxide conversion efficiency, injection response deviation and pressure difference change trend, the transient function deviation index and the long-term degradation accumulation index are calculated respectively. The transient function deviation index is used to characterize the short-term emission anomaly caused by control and thermal response lag, and the long-term degradation accumulation index is used to reflect the irreversible performance degradation of catalytic materials and filter structures under repeated thermal shocks. Based on the synergistic relationship between the transient functional deviation index and the long-term degradation cumulative index, a health evolution discrimination model for the exhaust gas treatment system is constructed. When an emission anomaly is detected as being caused by transient thermal mismatch, maintenance is triggered to suppress it. When the long-term degradation cumulative index shows a continuous increase and approaches the material tolerance limit, it is determined that the system has an accelerated life consumption trend, and predictive maintenance early warning information is generated.

[0016] In this embodiment of the invention, the process of collecting operational data of the ship's exhaust gas treatment system under different navigation stages, including main engine load, exhaust flow rate, inlet and outlet temperatures of the exhaust gas treatment system, pressure difference of the particulate filter, nitrogen oxide concentration, and injection control commands, and combining this with the catalyst structural dimensions, material heat capacity, and thermal conductivity coefficient, to calculate the catalyst's thermal inertia parameters and thermal diffusivity, and obtaining the thermal response parameter set of the exhaust gas treatment unit, is as follows: In the ship exhaust gas treatment system, high-temperature corrosion-resistant temperature sensors, differential pressure sensors and nitrogen oxide concentration sensors are arranged at the exhaust gas inlet and outlet, and the main engine load signal, exhaust flow signal and injection control command are obtained in the main engine control system. The temperature sensor is used to collect the inlet temperature of the exhaust gas treatment system. With outlet temperature The differential pressure sensor is used to collect the pressure difference before and after the particle trap. The nitrogen oxide concentration sensor is used to collect the nitrogen oxide concentration values ​​before and after exhaust gas treatment, while the main engine load and exhaust flow rate are used to characterize the intensity of changes in operating conditions under different navigation stages. The collected host load, exhaust flow, temperature, differential pressure, nitrogen oxide concentration, and injection control commands are all timestamped and time-sequentially aligned according to the preset sampling period to form the original operating data sequence of the exhaust gas treatment system. It should be noted that the sampling period is preferably 1s-5s. This time scale can cover typical transient conditions such as ship acceleration and deceleration, and can also avoid data redundancy from interfering with subsequent calculations. By combining the structural design parameters of the exhaust gas treatment system, the density and volume of different materials in the catalyst are obtained, and the heat capacity parameters of the catalyst are calculated by combining the specific heat capacity of the catalyst materials. For example, heat capacity parameter It can be calculated using the following formula: Where k represents the different material layers within the catalyst, including the carrier substrate layer, the catalyst coating, and necessary structural support layers, etc. Let be the density of the k-th layer material. Let k be the specific heat capacity of the k-th layer material. Let M be the volume corresponding to the k-th material layer, and M be the total number of material layers. It should be noted that the heat capacity parameter is used to characterize the amount of heat that the catalyst needs to absorb under temperature rise conditions. Its physical meaning is to reflect the thermal buffering capacity of the catalyst against transient heat input. This parameter directly affects the response rate of the catalytic reaction when the exhaust gas temperature changes rapidly. Based on the axial length of the catalyst and the thermal conductivity of the material, the characteristic time of heat diffusion along the axial direction of the catalyst is calculated to obtain the thermal diffusion time constant of the catalyst bed, which is used to characterize the hysteresis characteristics of the temperature field inside the catalyst from the inlet to the outlet. For example, the thermal diffusion time constant It can be calculated using the following formula: ,in The axial length of the catalyst. The thermal diffusivity; The thermal diffusivity can be calculated using the following formula: ,in The thermal conductivity coefficient, The average density of the material. The average specific heat capacity of the material; The thermal conductivity coefficient can be calculated using the following formula: ; It should be noted that the larger the thermal diffusion time constant, the more uneven the temperature distribution inside the catalyst, the longer the local high temperature or low temperature region lasts, and the more likely it is to produce a mismatch between thermal inertia and sensor measurement when there are frequent sudden changes in operating conditions. The heat capacity parameter and the thermal diffusion time constant are combined to construct a set of thermal response parameters for the exhaust gas treatment unit; It should be noted that the thermal response parameter set is used to describe the actual thermal response capability of the exhaust gas treatment system under different navigation stages and different operating conditions and rates of change.

[0017] In this embodiment of the invention, based on the thermal response parameter set, thermal inertia compensation and time reconstruction are performed on the collected exhaust temperature sequence to calculate the equivalent response temperature evolution reflecting the true internal thermal state. The process of introducing the exhaust temperature change rate and load change rate to construct a transient thermal shock intensity factor, used to characterize the degree of thermal shock under frequent acceleration and deceleration conditions, is as follows: Obtain the thermal capacity parameters of the catalyst from the thermal response parameter set. With thermal diffusion time constant And use it as a physical constraint parameter to describe the thermal inertial characteristics of the exhaust gas treatment unit; The collected inlet temperature of the exhaust gas treatment system With outlet temperature The exhaust temperature time series data are arranged according to a unified timestamp, where t represents the sampling time; It should be noted that the inlet and outlet temperatures reflect the temperature of the external boundary of the catalyst, rather than the actual reaction temperature inside the catalyst. Therefore, thermal inertia compensation and time reconstruction are required for the temperature time series. Based on the first-order thermal inertia model, thermal inertia compensation is applied to the exhaust temperature time series, and its equivalent reaction temperature is obtained. The evolutionary relationship can be expressed as: ,in The thermal diffusion time constant is used to characterize the hysteresis response of the catalyst's internal temperature to changes in the inlet temperature. The equivalent reaction temperature The specific expression of the evolutionary relationship is the basic form of the first-order thermal inertia model; Under discrete-time conditions, the equivalent reaction temperature The evolution relationship is numerically reconstructed to obtain the equivalent reaction temperature evolution sequence, which is used to approximately reflect the true thermal state of the catalyst after axial averaging. The equivalent reaction temperature under discrete-time conditions is described above. The process of numerically reconstructing the evolution relationship to obtain the equivalent reaction temperature evolution sequence is as follows: Equivalent reaction temperature The evolutionary relationship equation is discretized, and the Euler explicit difference method is used to approximate the differential equation: ,in This is the equivalent reaction temperature at the next moment. The equivalent reaction temperature at the current moment. The current exhaust gas inlet temperature. This represents the time step for discretization; Set the equivalent reaction temperature at the initial time t=0. It can usually be set according to the actual measured value or system design parameters; Obtain the exhaust gas inlet temperature The initial value; Iterate at each time step t, updating the numerical reconstruction using the formula described above. : ; Through continuous iteration, the equivalent reaction temperature sequence throughout the entire time series can be obtained: ; Under different working conditions It can be adjusted according to the actual data sampling frequency or the required resolution; Combined with exhaust temperature change rate The calculation method is as follows: ,in The time interval between adjacent sampling times; Synchronous calculation of host load change rate The calculation method is as follows: ,in The load level of the host at time t; After normalizing the exhaust temperature change rate and the load change rate, a weighted fusion was performed to construct the transient thermal shock intensity factor. The calculation formula is as follows: ,in, and These are preset proportional coefficients for the rate of change of exhaust temperature and the rate of change of load, respectively, used to balance the impact of changes in heat input and load on the degree of thermal shock. It should be noted that the transient thermal shock intensity factor is used to quantitatively describe the instantaneous thermal stress level borne by the exhaust gas treatment system during non-steady-state navigation phases such as frequent acceleration and deceleration. The larger the value, the more drastic the changes in the internal temperature field and reaction conditions of the system, and the higher the risk of thermal shock to the catalytic materials and filter structure. It should be noted that, and The settings should be tailored to the specific circumstances. For example, an expert-empowered approach could be adopted, where experts in relevant fields are invited to determine the pre-defined proportions for each indicator through professional opinion surveys and comprehensive evaluations. and The initial value can be 0.5, 0.5; The equivalent reaction temperature evolution and transient thermal shock intensity factor are used as important basic characteristics to distinguish between transient functional fluctuations and long-term material degradation, which are used to support the physical rationality and reliability of predictive maintenance decisions for the entire life cycle of ship exhaust gas treatment systems.

[0018] In this embodiment of the invention, based on the transient thermal shock intensity factor and the equivalent reaction temperature evolution, the thermal state evolution characteristics of the exhaust gas treatment system during operation are extracted, and a segmented thermal operating condition state vector including rapid thermal transition state and quasi-steady-state reaction state is constructed. The process of distinguishing between the strong transient operating condition stage and the continuous stable operating condition stage is as follows: Based on the evolution of equivalent reaction temperature With transient thermal shock intensity factor The thermal state changes of the exhaust gas treatment system during continuous operation are analyzed in conjunction to characterize the thermal conditions of the system under different navigation stages. A sliding window is used to divide the equivalent reaction temperature sequence on a continuous time axis, dividing the process into several adjacent time windows. Where k represents the time window number; It should be noted that the preferred time window length is 30s-120s. This window length can cover a typical ship acceleration or deceleration process, while avoiding misleading the judgment of the working condition due to a single transient peak. Within each time window, the average change in equivalent reaction temperature is calculated. Mean value of transient thermal shock intensity factor This is used to characterize the overall trend and impact intensity of thermal state changes within this time window; For example, the average variation of the equivalent reaction temperature It can be calculated using the following formula: ,in The equivalent reaction temperature at time point t. For time window The average value of the internal equivalent reaction temperature is calculated using the following formula: ; For example, the mean value of the transient thermal shock intensity factor It can be calculated using the following formula: ; The stability index of the rate of change of the equivalent reaction temperature within the time window is calculated and expressed as the standard deviation of the first difference of the equivalent reaction temperature: , ,in, Used to reflect the degree of fluctuation in the internal thermal state of the catalyst within this time window. This is the function for obtaining the standard deviation; The average change in the equivalent reaction temperature Mean value of transient thermal shock intensity factor and the standard deviation of the first difference of the equivalent reaction temperature These are combined as thermal state evolution features to construct the thermal condition state vector corresponding to the time window: ; Based on the thermal condition state vector and the preset thermal shock judgment threshold, the thermal condition type of each time window is classified and judged; the thermal shock judgment threshold includes the equivalent reaction temperature change amplitude threshold, the transient thermal shock intensity factor threshold, and the temperature change rate stability standard deviation threshold. When the average change of the equivalent reaction temperature is greater than or equal to the threshold of the equivalent reaction temperature change, the mean of the transient thermal shock intensity factor is greater than or equal to the threshold of the transient thermal shock intensity factor, and the standard deviation of the first difference of the equivalent reaction temperature is greater than or equal to the standard deviation threshold of the temperature change rate stability, the time window is determined to be in a rapid thermal transition state. This state is used to characterize the strong transient operating condition stage caused by frequent acceleration, deceleration or surge. This state indicates that the system is in a high temperature transient change stage, which is usually related to the ship's acceleration, deceleration or surge conditions. When the average change of the equivalent reaction temperature is less than the threshold of the equivalent reaction temperature change, the mean of the transient thermal shock intensity factor is less than the threshold of the transient thermal shock intensity factor, and the standard deviation of the first difference of the equivalent reaction temperature is less than the standard deviation threshold of the temperature change rate stability, the time window is determined to be in a quasi-steady-state reaction state. This state is used to characterize the continuous operation stage of the exhaust gas treatment system under relatively stable load conditions. This state usually indicates that the system is in a stable operating condition, the load is relatively stable, the temperature changes slowly, and the system operates relatively smoothly. It should be noted that by using the above-mentioned thermal condition segmentation method, the operation process can be segmented based on the physical thermal response characteristics without relying on a single emission index, thereby avoiding misjudging short-term performance fluctuations caused by transient thermal shock as long-term material degradation. The thermal condition type and thermal condition state vector corresponding to each time window are combined in chronological order to form a segmented state vector sequence of thermal conditions for the exhaust gas treatment system.

[0019] In this embodiment of the invention, based on the segmented state vector of thermal conditions, and combined with the nitrogen oxide conversion efficiency, injection response deviation, and pressure difference change trend, the transient functional deviation index and the long-term degradation accumulation index are calculated respectively. The transient functional deviation index is used to characterize short-term emission anomalies caused by control and thermal response lag, while the long-term degradation accumulation index reflects the irreversible performance degradation process of the catalytic material and filter structure under repeated thermal shocks. From the segmented state vector sequence of thermal conditions, obtain the thermal condition type identifier corresponding to each time window to distinguish between rapid thermal transition state and quasi-steady-state response state, and use it as the condition constraint condition for subsequent functional deviation and degradation assessment. Within each time window, the nitrogen oxide conversion efficiency is calculated based on the inlet and outlet nitrogen oxide concentration data. ; For example, nitrogen oxide conversion efficiency It can be calculated using the following formula: ,in, and These represent the average nitrogen oxide concentrations at the inlet and outlet of the exhaust gas treatment system within the k-th time window, respectively. Synchronously acquire injection control commands within this time window Compared with actual injection response Calculate the injection response deviation ; For example, injection response deviation It can be calculated using the following formula: ; Within the time window identified as a rapid thermal transition state, a transient functional deviation index is constructed based on the nitrogen oxide conversion efficiency and the injection response deviation. ; For example, transient function deviation index It can be calculated using the following formula: ,in , These represent preset proportional coefficients for nitrogen oxide conversion efficiency and injection response deviation, respectively, used to balance the impact of emission performance fluctuations and control response lag on transient function deviation; It should be noted that, , The settings should be tailored to the specific circumstances. For example, an expert-empowered approach could be adopted, where experts in relevant fields are invited to determine the pre-defined proportions for each indicator through professional opinion surveys and comprehensive evaluations. , The initial value can be 0.5, 0.5; It should be noted that the transient performance deviation index is used to characterize short-term emission anomalies caused by thermal inertia, control response lag, or insufficient injection adaptation under conditions of strong thermal shock. This index has obvious operating condition correlation and recoverability. Within the time window determined to be a quasi-steady-state reaction state, the trend of pressure difference changes before and after the particle trap is considered. It is represented by the difference in the average pressure difference between adjacent time windows; For example, the trend of pressure difference before and after the particulate filter. It can be calculated using the following formula: ; By combining the aforementioned pressure difference change trend with the long-term evolution characteristics of nitrogen oxide conversion efficiency, a long-term degradation cumulative index is constructed. ; For example, the long-term cumulative degradation index It can be calculated using the following formula: ,in For historical time window time index variables, , , These are preset proportional coefficients for the changes in nitrogen oxide conversion efficiency and pressure difference, used to reflect the impact of continuous decline in catalytic efficiency and irreversible blockage of the filter structure on system lifespan. It should be noted that, , The settings should be tailored to the specific circumstances. For example, an expert-empowered approach could be adopted, where experts in relevant fields are invited to determine the pre-defined proportions for each indicator through professional opinion surveys and comprehensive evaluations. , The initial value can be 0.5, 0.5; It should be noted that the long-term degradation accumulation index effectively suppresses the short-term fluctuation interference caused by transient thermal shock by integrating the performance degradation over time under quasi-steady-state conditions, thus more accurately reflecting the irreversible degradation processes such as the decline in the activity of catalytic materials and the accumulation of ash in the particle trapping structure. The transient functional deviation index and the long-term degradation cumulative index are associated with the corresponding thermal condition segmented state vectors to form a functional deviation-degradation evolution feature set of the exhaust gas treatment system under different operating conditions.

[0020] In this embodiment of the invention, a health evolution discrimination model for the exhaust gas treatment system is constructed based on the synergistic relationship between the transient functional deviation index and the long-term degradation accumulation index. When an emission anomaly is detected as being caused by transient thermal mismatch, maintenance is suppressed. When the long-term degradation accumulation index shows a continuous increase and approaches the material tolerance limit, it is determined that the system has an accelerated lifespan consumption trend. The process of generating predictive maintenance early warning information is as follows: The health evolution discrimination model consists of an input layer, a discrimination layer, and a decision layer; The input layer uses the comprehensive health status representation vector of the exhaust gas treatment system as the status input. The discriminant layer is used to distinguish different health evolution states by analyzing the changing characteristics of the comprehensive health status representation vector on the time axis; The decision-making layer is used to output maintenance decision constraint logic; According to transient function deviation index With long-term cumulative degradation index A dual-index health evolution feature space is constructed for the exhaust gas treatment system during operation to characterize the co-evolutionary relationship between system functional fluctuations and structural degradation. A health status discriminant function is introduced into the dual-index health evolution feature space. Its form is: ,in This represents the comprehensive health status representation vector of the exhaust gas treatment system within the k-th time window. Based on historical operating data and material tolerance characteristics, a transient function deviation threshold range is preset in the dual-index feature space. With long-term degradation risk threshold Among them, the transient function deviation from the threshold range is used to define the acceptable range of short-term emission anomalies. This represents the maximum acceptable range for short-term emission anomalies. The threshold represents the minimum acceptable range for short-term emission anomalies, while the long-term degradation risk threshold is used to characterize the state where the activity of the catalytic material declines or the filter structure becomes clogged and approaches its tolerance limit. When transient functional deviation from the index is detected within a consecutive number of time windows The levels continued to rise and all exceeded the maximum acceptable range for short-term emission anomalies. However, the long-term cumulative degradation index Less than or equal to the preset long-term degradation risk threshold At that time, it was determined that the current emission anomaly was caused by transient thermal mismatch due to thermal inertia and control response lag; Under the above-mentioned judgment conditions, the health evolution discrimination model enters the maintenance suppression state, suppresses the maintenance trigger signal, and outputs the operation status interpretation information simultaneously to indicate that the anomaly belongs to a recoverable functional deviation rather than an irreversible structural degradation. When the long-term degradation cumulative index is detected within multiple consecutive quasi-steady-state reaction time windows... It exhibits a monotonically increasing trend, and the value is greater than the preset long-term degradation risk threshold. At that time, it was determined that the exhaust gas treatment system showed a trend of accelerated lifespan consumption; It should be noted that the continuous increase in the long-term degradation accumulation index reflects the irreversible decline in the activity of the catalytic material under repeated thermal shock, or the continuous accumulation of ash inside the particulate trap leading to a decrease in the effective volume of the filter structure, a process that is irreversible. After determining that there is an accelerated lifespan consumption trend, the health evolution discrimination model generates predictive maintenance warning information and outputs the warning information together with the corresponding degradation stage label, remaining lifespan trend and suggested maintenance window to guide maintenance personnel to carry out targeted maintenance operations at the appropriate time.

[0021] This invention, by introducing a thermal inertia compensation and time reconstruction mechanism, maps the exhaust temperature and emission data obtained by limited sensors into an equivalent reaction temperature evolution that can characterize the actual heating state inside the catalyst. Combined with the transient thermal shock intensity factor and the segmented state vector of thermal conditions, it achieves an effective distinction between strong transient conditions and quasi-steady-state conditions. It constructs a collaborative discrimination system of transient functional deviation index and long-term degradation accumulation index, thereby enabling accurate differentiation between short-term emission fluctuations caused by control lag and thermal response mismatch and irreversible material degradation caused by repeated thermal shock accumulation under complex navigation conditions.

[0022] This invention avoids the risk of misjudgment caused by relying solely on experience-based judgments based on a single emission or pressure difference index, significantly reduces the probability of triggering unnecessary maintenance, and can provide early warnings before the accelerated consumption of catalytic materials and filter structures, achieving truly predictive maintenance throughout the entire life cycle. This improves the operational reliability, maintenance decision accuracy, and engineering applicability of ship exhaust gas treatment systems in real navigation environments.

[0023] Example 2: This example introduces a predictive maintenance system for the entire lifecycle of a ship's exhaust gas treatment system, such as... Figure 2As shown, it includes a runtime perception modeling module, a thermal response reconstruction module, a thermal condition segmentation module, a degradation dual-index assessment module, and a health judgment decision module; The perception modeling module is used to collect operational data of the ship's exhaust gas treatment system at different navigation stages. Combined with the catalyst structure size, material heat capacity and thermal conductivity, the thermal inertia parameters and thermal diffusion time constant of the catalyst are calculated to obtain the thermal response parameter set of the exhaust gas treatment unit. The thermal response reconstruction module is used to perform thermal inertia compensation and time reconstruction on the collected exhaust temperature sequence according to the thermal response parameter set, calculate the equivalent reaction temperature evolution, and introduce the exhaust temperature change rate and load change rate to construct the transient thermal shock intensity factor. The thermal condition segmentation module is used to extract the thermal state evolution characteristics of the exhaust gas treatment system during operation based on the transient thermal shock intensity factor and the equivalent reaction temperature evolution, and to construct a thermal condition segmentation state vector that includes rapid thermal transition state and quasi-steady-state reaction state. The degradation dual-index assessment module is used to calculate the transient functional deviation index and the long-term degradation cumulative index based on the segmented state vector of thermal conditions, combined with the nitrogen oxide conversion efficiency, injection response deviation and pressure difference change trend. The health assessment and decision-making module is used to construct a health evolution assessment model for the exhaust gas treatment system based on the synergistic relationship between the transient functional deviation index and the long-term degradation cumulative index. When an emission anomaly is detected as being caused by transient thermal mismatch, maintenance is triggered to suppress it. When the long-term degradation cumulative index shows a continuous increase and approaches the material tolerance limit, it is determined that the system has an accelerated life consumption trend, and predictive maintenance early warning information is generated.

[0024] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0025] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0026] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0027] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the system and method described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0028] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways.

[0029] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A predictive maintenance method for the entire lifecycle of a ship's exhaust gas treatment system, characterized in that: The process includes the following steps: collecting operational data of the ship's exhaust gas treatment system at different navigation stages, and combining the catalyst structure dimensions, material heat capacity and thermal conductivity coefficient to calculate the catalyst thermal inertia parameters and thermal diffusion time constant, thereby obtaining the thermal response parameter set of the exhaust gas treatment unit; Based on the thermal response parameter set, thermal inertia compensation and time reconstruction are performed on the collected exhaust temperature sequence to calculate the equivalent reaction temperature evolution, and the exhaust temperature change rate and load change rate are introduced to construct the transient thermal shock intensity factor. Based on the transient thermal shock intensity factor and the evolution of the equivalent reaction temperature, the thermal state evolution characteristics of the exhaust gas treatment system during operation are extracted, and a segmented thermal condition state vector including rapid thermal transition state and quasi-steady-state reaction state is constructed. Based on the segmented state vector of thermal conditions, the transient function deviation index and the long-term degradation accumulation index are calculated by combining the nitrogen oxide conversion efficiency, injection response deviation and pressure difference change trend. Based on the synergistic relationship between the transient functional deviation index and the long-term degradation cumulative index, a health evolution discrimination model for the exhaust gas treatment system is constructed. When an emission anomaly is detected as being caused by transient thermal mismatch, maintenance is triggered to suppress it. When the long-term degradation cumulative index shows a continuous increase and approaches the material tolerance limit, it is determined that the system has an accelerated life consumption trend, and predictive maintenance early warning information is generated.

2. The predictive maintenance method for the entire life cycle of a ship exhaust gas treatment system according to claim 1, characterized in that: The process of collecting operational data of the ship's exhaust gas treatment system at different navigation stages, and combining this data with the catalyst's structural dimensions, material heat capacity, and thermal conductivity to calculate the catalyst's thermal inertia parameters and thermal diffusion time constant, thus obtaining the thermal response parameter set of the exhaust gas treatment unit, is as follows: In the ship exhaust gas treatment system, high-temperature corrosion-resistant temperature sensors, differential pressure sensors and nitrogen oxide concentration sensors are arranged at the exhaust gas inlet and outlet, and the main engine load signal, exhaust flow signal and injection control command are obtained in the main engine control system. The temperature sensor is used to collect the inlet temperature of the exhaust gas treatment system. With outlet temperature The differential pressure sensor is used to collect the pressure difference before and after the particle trap. The nitrogen oxide concentration sensor is used to collect nitrogen oxide concentration values ​​before and after exhaust gas treatment; The collected host load, exhaust flow, temperature data, differential pressure data, nitrogen oxide concentration data and injection control commands are unified with timestamps and time-aligned according to the preset sampling period to form the original operating data sequence of the exhaust gas treatment system. By combining the structural design parameters of the exhaust gas treatment system, the density and volume of different materials in the catalyst are obtained, and the heat capacity parameters of the catalyst are calculated by combining the specific heat capacity of the catalyst materials. Based on the axial length of the catalyst and the thermal conductivity of the material, the characteristic time of heat diffusion along the axial direction of the catalyst is calculated, and the thermal diffusion time constant of the catalyst bed is obtained. The heat capacity parameter and the thermal diffusion time constant are combined to construct a set of thermal response parameters for the exhaust gas treatment unit.

3. The predictive maintenance method for the entire life cycle of a ship exhaust gas treatment system according to claim 2, characterized in that: Based on the aforementioned thermal response parameter set, the process of performing thermal inertia compensation and time reconstruction on the collected exhaust temperature sequence, calculating the equivalent reaction temperature evolution, and introducing the exhaust temperature change rate and load change rate to construct the transient thermal shock intensity factor is as follows: From the set of thermal response parameters, the thermal capacity parameters and thermal diffusion time constant of the catalyst are obtained and used as physical constraint parameters to describe the thermal inertial characteristics of the exhaust gas treatment unit. The collected inlet temperature of the exhaust gas treatment system With outlet temperature The exhaust temperature time series data is formed by arranging the data according to a unified timestamp; Based on the first-order thermal inertia model, thermal inertia compensation is applied to the exhaust temperature time series, and its equivalent reaction temperature is obtained. The evolutionary relationship can be expressed as: ,in The thermal diffusion time constant; Under discrete-time conditions, the equivalent reaction temperature The evolution relationship was numerically reconstructed to obtain the equivalent reaction temperature evolution sequence; Combined with exhaust temperature change rate The calculation method is as follows: ,in The time interval between adjacent sampling times; Synchronous calculation of host load change rate The calculation method is as follows: ,in The load level of the host at time t; After normalizing the exhaust temperature change rate and the load change rate, a weighted fusion was performed to construct the transient thermal shock intensity factor.

4. The predictive maintenance method for the entire life cycle of a ship exhaust gas treatment system according to claim 3, characterized in that: Based on the transient thermal shock intensity factor and the evolution of the equivalent reaction temperature, the thermal state evolution characteristics of the exhaust gas treatment system during operation are extracted. The process of constructing a segmented thermal condition state vector that includes rapid thermal transition states and quasi-steady-state reaction states is as follows: Based on the equivalent reaction temperature evolution and the transient thermal shock intensity factor, the thermal state changes of the exhaust gas treatment system during continuous operation are jointly analyzed. A sliding window is used to divide the equivalent reaction temperature sequence on a continuous time axis, dividing the process into several adjacent time windows. Where k represents the time window number; Within each time window, the average change in equivalent reaction temperature and the mean value of transient thermal shock intensity factor are calculated respectively. The stability index of the rate of change of the equivalent reaction temperature within the time window is calculated and expressed by the standard deviation of the first difference of the equivalent reaction temperature; The average variation of the equivalent reaction temperature, the mean of the transient thermal shock intensity factor, and the standard deviation of the first difference of the equivalent reaction temperature are combined as thermal state evolution characteristic quantities to construct the thermal condition state vector corresponding to the time window. Based on the thermal condition state vector and the preset thermal shock judgment threshold, the thermal condition type of each time window is classified and judged; the thermal shock judgment threshold includes the equivalent reaction temperature change amplitude threshold, the transient thermal shock intensity factor threshold, and the temperature change rate stability standard deviation threshold. When the average change of equivalent reaction temperature is greater than or equal to the threshold of equivalent reaction temperature change, the mean of transient thermal shock intensity factor is greater than or equal to the threshold of transient thermal shock intensity factor, and the standard deviation of the first difference of equivalent reaction temperature is greater than or equal to the standard deviation threshold of temperature change rate stability, the time window is determined to be in a rapid thermal transition state. When the average change of equivalent reaction temperature is less than the threshold of equivalent reaction temperature change, the mean of transient thermal shock intensity factor is less than the threshold of transient thermal shock intensity factor, and the standard deviation of the first difference of equivalent reaction temperature is less than the standard deviation threshold of temperature change rate stability, the time window is determined to be in a quasi-steady-state reaction state. The thermal condition type and thermal condition state vector corresponding to each time window are combined in chronological order to form a segmented state vector sequence of thermal conditions for the exhaust gas treatment system.

5. The predictive maintenance method for the entire life cycle of a ship exhaust gas treatment system according to claim 4, characterized in that: Based on the segmented state vector of thermal conditions, and combined with the nitrogen oxide conversion efficiency, injection response deviation, and pressure difference change trend, the process of calculating the transient performance deviation index and the long-term degradation accumulation index is as follows: From the segmented state vector sequence of thermal conditions, obtain the thermal condition type identifier corresponding to each time window to distinguish between rapid thermal transition state and quasi-steady-state reaction state. Within each time window, the nitrogen oxide conversion efficiency is calculated based on the inlet and outlet nitrogen oxide concentration data; Simultaneously acquire the injection control command and actual injection response within the time window, and calculate the injection response deviation; Within the time window determined to be a rapid thermal transition state, a transient functional deviation index is constructed based on the nitrogen oxide conversion efficiency and injection response deviation. Within the time window determined to be a quasi-steady-state reaction state, the difference between the average pressure difference of adjacent time windows is used to represent the trend of pressure difference changes before and after the particle trap. By combining the aforementioned pressure difference change trend with the long-term evolution characteristics of nitrogen oxide conversion efficiency, a long-term degradation cumulative index is constructed. The transient functional deviation index and the long-term degradation cumulative index are associated with the corresponding thermal condition segmented state vectors to form a functional deviation-degradation evolution feature set of the exhaust gas treatment system under different operating conditions.

6. The predictive maintenance method for the entire life cycle of a ship exhaust gas treatment system according to claim 5, characterized in that: Based on the synergistic relationship between the transient functional deviation index and the long-term degradation cumulative index, a health evolution discrimination model for the exhaust gas treatment system is constructed. When an emission anomaly is detected as being caused by transient thermal mismatch, suppression maintenance is triggered. When the long-term degradation cumulative index shows a continuous increase and approaches the material tolerance limit, it is determined that the system has an accelerated lifespan consumption trend. The process of generating predictive maintenance early warning information is as follows: Based on the transient functional deviation index and the long-term degradation cumulative index, a dual-index health evolution characteristic space of the exhaust gas treatment system during operation is constructed. In the dual-index health evolution feature space, a health status discrimination function is introduced; Based on historical operating data and material tolerance characteristics, a transient function deviation threshold range is preset in the dual-index feature space. With long-term degradation risk threshold ; When transient functional deviation from the index is detected within a consecutive number of time windows The levels continued to rise and all exceeded the maximum acceptable range for short-term emission anomalies. However, the long-term cumulative degradation index Less than or equal to the preset long-term degradation risk threshold At that time, it was determined that the current emission anomaly was caused by transient thermal mismatch due to thermal inertia and control response lag; Under the above-mentioned conditions, the health evolution discrimination model enters the maintenance suppression state and suppresses the maintenance trigger signal. When the long-term degradation cumulative index is detected within multiple consecutive quasi-steady-state reaction time windows... It exhibits a monotonically increasing trend, and the value is greater than the preset long-term degradation risk threshold. At that time, it was determined that the exhaust gas treatment system showed a trend of accelerated lifespan consumption; After determining that there is a trend of accelerated lifespan depletion, the health evolution discriminant model generates predictive maintenance early warning information.

7. A predictive maintenance system for the entire lifecycle of a ship exhaust gas treatment system, used to implement the predictive maintenance method for the entire lifecycle of a ship exhaust gas treatment system as described in any one of claims 1-6, characterized in that: It includes a runtime perception modeling module, a thermal response reconstruction module, a thermal condition segmentation module, a degradation dual-index assessment module, and a health judgment decision module; The perception modeling module is used to collect operational data of the ship's exhaust gas treatment system at different navigation stages. Combined with the catalyst structure size, material heat capacity and thermal conductivity, the thermal inertia parameters and thermal diffusion time constant of the catalyst are calculated to obtain the thermal response parameter set of the exhaust gas treatment unit. The thermal response reconstruction module is used to perform thermal inertia compensation and time reconstruction on the collected exhaust temperature sequence according to the thermal response parameter set, calculate the equivalent reaction temperature evolution, and introduce the exhaust temperature change rate and load change rate to construct the transient thermal shock intensity factor. The thermal condition segmentation module is used to extract the thermal state evolution characteristics of the exhaust gas treatment system during operation based on the transient thermal shock intensity factor and the equivalent reaction temperature evolution, and to construct a thermal condition segmentation state vector that includes rapid thermal transition state and quasi-steady-state reaction state. The degradation dual-index assessment module is used to calculate the transient functional deviation index and the long-term degradation cumulative index based on the segmented state vector of thermal conditions, combined with the nitrogen oxide conversion efficiency, injection response deviation and pressure difference change trend. The health assessment and decision-making module is used to construct a health evolution assessment model for the exhaust gas treatment system based on the synergistic relationship between the transient functional deviation index and the long-term degradation cumulative index. When an emission anomaly is detected as being caused by transient thermal mismatch, maintenance is triggered to suppress it. When the long-term degradation cumulative index shows a continuous increase and approaches the material tolerance limit, it is determined that the system has an accelerated life consumption trend, and predictive maintenance early warning information is generated.

Images