Intelligent engine oil sensor early warning system with self-diagnosis function

Abstract

The invention relates to the technical field of engine oil sensor early warning, and discloses an intelligent engine oil sensor early warning system with a self-diagnosis function, and the system is characterized in that a data modeling module firstly divides to-be-monitored sub-parts, and then fuses historical and real-time data to construct a fault feature library; the risk assessment module determines an early warning level based on the current data deviation degree and a preset corresponding relation; and the dynamic adjustment module compares the current data with the historical data, adjusts the early warning level and determines the acquisition frequency, and finally forms a closed-loop monitoring system from data modeling, risk assessment to dynamic adjustment, thereby realizing accurate early warning and efficient diagnosis. Through cooperation of the data modeling module, the risk assessment module, the dynamic adjustment module and the like, accurate monitoring, intelligent early warning and dynamic optimization of the operation state of the engine oil system are achieved, the fault pre-judgment capacity is effectively improved, the equipment maintenance cost is reduced, and safe and stable operation of an engine is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of engine oil sensor warning technology, and more specifically, to an intelligent engine oil sensor warning system with self-diagnostic function. Background Technology

[0002] The oil sensor is a core component for monitoring the condition of the engine lubrication system. It provides data support for the reliable operation of the engine by collecting parameters such as oil pressure, temperature, level, and quality in real time. With the development of intelligent and complex engines, higher requirements are placed on the accuracy and real-time performance of fault warnings for the oil system.

[0003] However, traditional oil sensor monitoring systems have significant shortcomings: First, they can only monitor a single parameter independently, lacking the ability to fuse and analyze multi-dimensional data and comprehensively assess the overall system operating status, making it difficult to capture the correlation and fault characteristics between parameters. Second, their reliance on fixed threshold warning mechanisms prevents them from dynamically learning from historical operating data and predicting potential faults, leading to missed detections or delayed responses to early abnormal signals. Third, their fixed data acquisition frequency, without adaptive adjustments based on engine operating conditions (such as idling and high speed) and fault risk levels, easily results in missed key abnormal data or redundant invalid data under complex operating conditions. Fourth, they lack self-diagnosis and fault tracing capabilities; when the sensor itself exhibits abnormal signals, it cannot effectively identify and switch redundant monitoring strategies, leading to decreased warning reliability. These problems directly affect the accuracy of engine fault prediction, increasing maintenance costs and safety hazards.

[0004] Therefore, there is an urgent need to design an intelligent oil sensor early warning system with self-diagnostic function. By constructing an operational data-fault feature database, dynamically adjusting the risk assessment model, and adaptively optimizing the acquisition strategy, the system can solve the technical bottlenecks of traditional systems in terms of multi-parameter fusion analysis, fault prediction accuracy, dynamic monitoring capability, and self-diagnostic reliability. Summary of the Invention

[0005] In view of this, the present invention proposes an intelligent oil sensor early warning system with self-diagnostic function, which aims to solve the technical bottlenecks of traditional systems in terms of multi-parameter fusion analysis, fault prediction accuracy, dynamic monitoring capability and self-diagnostic reliability.

[0006] This invention proposes an intelligent oil sensor early warning system with self-diagnostic function, comprising:

[0007] The data modeling module is used to determine the part to be monitored, divide the part to be monitored into several sub-parts to be monitored, obtain the historical operation data and real-time operation data of each sub-part to be monitored, and establish an operation data-fault feature database after matching the historical operation data and real-time operation data of each sub-part to be monitored.

[0008] The risk assessment module is used to receive the current operating data of each of the monitored sub-parts, and to determine the risk level of historical faults based on the frequency and severity of faults in the historical operating data; it also pre-sets the correspondence between the degree of data anomaly and the warning level, and determines the current warning level based on the degree to which the current operating data deviates from the normal range, combined with the correspondence between the degree of data anomaly and the warning level.

[0009] The dynamic adjustment module is used to compare the current operating data of each of the monitored sub-parts with the historical operating data, determine the current fault risk level based on the comparison results, adjust the current warning level based on the determined current fault risk level to determine the final warning level, and determine the sampling frequency for fault diagnosis based on the final warning level.

[0010] Furthermore, when determining the sampling frequency for fault diagnosis based on the final warning level, the dynamic adjustment module further includes:

[0011] The dynamic adjustment module determines the operating condition level based on different engine operating condition information, including idling condition, low speed condition, medium speed condition, and high speed condition.

[0012] A pre-defined correspondence between operating condition levels and acquisition frequency adjustment coefficients is established. Based on the operating condition levels, acquisition frequency adjustment coefficients are selected from the correspondence to adjust the acquisition frequency during fault diagnosis, thereby obtaining the final acquisition frequency.

[0013] Based on the final acquisition frequency, various sensors are controlled to acquire data from the sub-parts to be monitored, and the acquired data is transmitted to the data modeling module and risk assessment module for continuous optimization of the operation data-fault feature database and updating of fault risk level and early warning level.

[0014] Furthermore, the dynamic adjustment module categorizes historical fault risk levels into potential fault risk, moderate fault risk, and severe fault risk;

[0015] When the dynamic adjustment module determines that there is a potential fault risk in the monitored sub-part, it sets a dynamic monitoring threshold and increases the data collection frequency based on the historical operating data and real-time data fluctuation range of the potential risk area.

[0016] When the dynamic adjustment module determines that there is a moderate risk of failure in the monitored sub-part, it will activate the preset maintenance plan based on the type and severity of the failure and adjust the key parameters of data acquisition.

[0017] When the dynamic adjustment module determines that there is a serious risk of failure in the monitored sub-part, it immediately triggers the emergency maintenance process based on the location of the failure and the possible cause, and simultaneously notifies relevant personnel to handle the situation.

[0018] Furthermore, when determining the current warning level, the risk assessment module also includes: pre-setting weighting coefficients for different types of sensor data, including pressure data weighting coefficients, temperature data weighting coefficients, liquid level data weighting coefficients, and quality data weighting coefficients;

[0019] Based on the degree to which each type of data in the current operating data deviates from the normal range and the corresponding weighting coefficient, a comprehensive outlier value is calculated.

[0020] The current warning level is determined by comparing the comprehensive anomaly value with a pre-set anomaly level range, wherein the anomaly level range includes a first anomaly level range, a second anomaly level range, and a third anomaly level range.

[0021] When the overall outlier value falls within the first outlier level range, the current warning level is determined to be a Level 1 warning.

[0022] When the overall outlier value falls within the second outlier level range, the current warning level is determined to be a Level II warning.

[0023] When the comprehensive outlier value falls within the third outlier level range, the current warning level is determined to be a Level III warning.

[0024] Furthermore, when establishing the operational data-fault feature database, the data modeling module also includes: preprocessing the acquired historical operational data and real-time operational data, the preprocessing including data cleaning and normalization, analyzing the preprocessed data, and classifying data with similar characteristics into the same category;

[0025] A corresponding fault feature description is established for each category. The fault feature description includes fault type, fault occurrence probability and fault manifestation. The data category and the corresponding fault feature description are associated and stored in the operation data-fault feature database.

[0026] Furthermore, when the dynamic adjustment module adjusts the current warning level according to the current fault risk level, it includes:

[0027] A risk level adjustment coefficient table is pre-defined, which includes adjustment coefficients for potential failure risks, adjustment coefficients for moderate failure risks, and adjustment coefficients for severe failure risks.

[0028] Select the corresponding adjustment coefficient based on the determined current fault risk level, multiply the current warning level by the adjustment coefficient to obtain the adjusted warning level. If the adjusted warning level exceeds the preset maximum warning level, then set it as the maximum warning level.

[0029] Furthermore, the system also includes an alarm module, which is electrically connected to the dynamic adjustment module;

[0030] When the final warning level reaches Level 1, the alarm module emits a low-frequency warning sound and provides a slow flashing indicator light.

[0031] When the final warning level reaches Level II, the alarm module emits a medium-frequency warning sound, the indicator light flashes rapidly, and a warning message is sent to the vehicle's dashboard.

[0032] When the final warning level reaches Level 3, the alarm module emits a high-frequency warning sound, the indicator light stays on, and an emergency warning message is sent to the vehicle's dashboard. At the same time, an alarm SMS message is sent to a preset user terminal.

[0033] Furthermore, the risk assessment module is also used to: obtain the cumulative running time of the engine, and pre-set different cumulative running time intervals including a first time interval, a second time interval and a third time interval, with the corresponding fault risk correction coefficients being the first fault risk correction coefficient, the second fault risk correction coefficient and the third fault risk correction coefficient;

[0034] Based on the range of the engine's cumulative operating time, the corresponding fault risk correction coefficient is selected to correct the historical fault risk level.

[0035] The revised historical fault risk levels are integrated with the current fault risk levels for analysis, thereby optimizing the determination of the final warning level.

[0036] Furthermore, after determining the sampling frequency for fault diagnosis based on the final warning level, the dynamic adjustment module also includes:

[0037] The upper and lower limits of the sampling frequency are preset; when the final sampling frequency after adjustment is greater than the upper limit, the final sampling frequency is set to the upper limit frequency value; when the final sampling frequency is less than the lower limit, the final sampling frequency is set to the lower limit frequency value.

[0038] If, in multiple consecutive data acquisitions, the operating data of each monitored sub-part are within the normal range and the final warning level is the lowest level, then the final acquisition frequency will be adjusted by reducing it according to the preset frequency attenuation coefficient, but the adjusted frequency will not be lower than the lower limit frequency value.

[0039] Furthermore, the system also includes a maintenance suggestion generation module, which is electrically connected to the dynamic adjustment module;

[0040] When the dynamic adjustment module determines that there is a moderate or severe fault risk, the maintenance suggestion generation module generates detailed maintenance suggestions based on the fault feature information in the operation data-fault feature database, combined with the current fault risk level and fault type.

[0041] The maintenance recommendations include suggestions for replacing faulty components, instructions for repair procedures, and suggestions for adjusting maintenance cycles. These recommendations will be sent to the vehicle information display screen and the user's mobile application.

[0042] Compared with the prior art, the beneficial effects of the present invention are as follows: by constructing a refined operating data-fault feature database through the data modeling module, a structured analysis of multi-dimensional data of the engine oil system and in-depth mining of fault features are realized, providing a data foundation for accurate fault diagnosis;

[0043] The risk assessment module is based on a dual assessment mechanism of historical data and real-time anomaly level, which breaks through the limitations of traditional fixed threshold early warning and significantly improves the scientific nature of fault risk level determination and the real-time nature of early warning response.

[0044] The dynamic adjustment module dynamically optimizes the early warning level by comparing and analyzing real-time data with historical data, and intelligently matches the data collection frequency with fault risk. This avoids redundant invalid data and ensures the accurate capture of key abnormal signals, forming a closed-loop self-diagnostic system of "data modeling - risk assessment - dynamic optimization". Ultimately, it realizes intelligent and adaptive monitoring of the engine oil system's operating status, effectively reducing the fault omission rate, improving maintenance efficiency, and ensuring the safe and reliable operation of the engine. Attached Figure Description

[0045] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0046] Figure 1 This is a first functional block diagram of an intelligent oil sensor early warning system with self-diagnostic function provided in an embodiment of the present invention;

[0047] Figure 2 This is a second functional block diagram of an intelligent oil sensor early warning system with self-diagnostic function provided in an embodiment of the present invention. Detailed Implementation

[0048] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0049] Reference Figure 1 In some embodiments of this application, an intelligent oil sensor warning system with self-diagnostic function includes:

[0050] Reference Figure 1 As shown in some embodiments of this application, an intelligent oil sensor early warning system with self-diagnostic function includes:

[0051] The data modeling module is used to determine the part to be monitored, divide the part to be monitored into several sub-parts to be monitored, obtain the historical and real-time operating data of each sub-part, and establish an operating data-fault feature database after matching the historical and real-time operating data of each sub-part.

[0052] The risk assessment module receives current operational data from each monitored sub-part, models the frequency and severity of faults in historical operational data (e.g., calculates the prior probability of each fault type using a Bayesian probability model, and classifies risk levels based on the severity of the fault consequences), and determines the historical fault risk level; it also pre-sets the correspondence between the degree of data anomaly and the warning level, and determines the current warning level based on the degree to which the current operational data deviates from the normal range, combined with the correspondence between the degree of data anomaly and the warning level.

[0053] The dynamic adjustment module is used to compare the current operating data of each monitored sub-part with the historical operating data, determine the current fault risk level based on the comparison results, adjust the current warning level based on the determined current fault risk level, and determine the final warning level. The sampling frequency during fault diagnosis is determined based on the final warning level.

[0054] The above embodiments construct a refined operational data-fault feature database through a data modeling module, enabling structured analysis and in-depth mining of multi-dimensional data from the engine oil system, providing a data foundation for accurate fault diagnosis. The risk assessment module, based on a dual assessment mechanism of historical data and real-time anomaly levels, overcomes the limitations of traditional fixed-threshold warnings, significantly improving the scientific rigor of fault risk level determination and the real-time nature of warning responses. The dynamic adjustment module dynamically optimizes warning levels through comparative analysis of real-time and historical data, and intelligently matches data collection frequency to fault risk, avoiding redundant invalid data while ensuring accurate capture of key abnormal signals. This forms a closed-loop self-diagnostic system of "data modeling-risk assessment-dynamic optimization," ultimately achieving intelligent and adaptive monitoring of the engine oil system's operating status, effectively reducing the rate of missed fault detection, improving maintenance efficiency, and ensuring the safe and reliable operation of the engine.

[0055] Specifically, when determining the data collection frequency for fault diagnosis based on the final warning level, the dynamic adjustment module also includes:

[0056] The dynamic adjustment module determines the operating condition level based on different engine operating condition information, including idling condition, low speed condition, medium speed condition, and high speed condition.

[0057] The correspondence between operating condition level and acquisition frequency adjustment coefficient is preset. Based on the operating condition level, the acquisition frequency adjustment coefficient is selected from the correspondence to adjust the acquisition frequency during fault diagnosis and obtain the final acquisition frequency.

[0058] Based on the final acquisition frequency, various sensors are controlled to acquire data from the sub-parts to be monitored, and the acquired data is transmitted to the data modeling module and risk assessment module for continuous optimization of the operation data-fault characteristic database and updating of fault risk level and early warning level.

[0059] Specifically, the quantitative classification standard for operating condition levels is as follows:

[0060] Idle condition: Engine speed ≤ 800 rpm and vehicle speed ≤ 0 km / h, duration ≥ 10s;

[0061] Low-speed operating conditions: 800rpm < engine speed ≤ 1500rpm, and vehicle speed ≤ 30km / h;

[0062] Medium-speed operating conditions: 1500rpm < engine speed ≤ 3000rpm, or 30km / h < vehicle speed ≤ 80km / h;

[0063] High-speed operating conditions: Engine speed > 3000 rpm, or vehicle speed > 80 km / h;

[0064] Special operating conditions: rapid acceleration (acceleration > 2 m / s²) 2 Rapid deceleration (deceleration > -3m / s) 2 ), heavy load (load factor > 0.8).

[0065] Specifically, a data importance scoring system is established (pressure change rate > temperature gradient > quality index > liquid level fluctuation). When transmission bandwidth is insufficient, data items with high scores are transmitted first, while less important data is transmitted using compression or buffering and retransmission.

[0066] Specifically, when determining the fault diagnosis acquisition frequency, the dynamic adjustment module first divides the operating conditions into idling, low / medium / high speed, and special operating conditions based on parameters such as engine speed and vehicle speed. It presets the mapping relationship between each operating condition level and the acquisition frequency adjustment coefficient (e.g., the adjustment coefficient for high speed is 1.5, and for idling it is 0.7). Combined with the base frequency determined by the final warning level, it calculates the final acquisition frequency (the formula includes a fault risk correction factor) and controls the sensors to acquire data according to multiple strategies (Kalman filtering, priority transmission, and redundancy verification). At the same time, it transmits the real-time data to the modeling and evaluation module. Through mechanisms such as sliding time window incremental learning and fault feature weight enhancement, it continuously optimizes the operating data-fault feature database, realizing dynamic adaptive adjustment of the acquisition frequency according to the operating conditions and fault risks. This reduces invalid data redundancy while improving the accuracy of key abnormal signal capture and gives the system fault feature database the ability to evolve.

[0067] Understandably, the dynamic adjustment module precisely matches the engine operating condition level with the acquisition frequency adjustment coefficient, and dynamically corrects the acquisition strategy based on the fault risk level. This enables the data acquisition frequency to be adaptively adjusted according to the complexity of the operating conditions and the degree of fault risk. Under complex operating conditions such as high speed and heavy load, the acquisition frequency is increased to ensure the accurate capture of key abnormal signals, while the frequency is reduced to reduce resource consumption under low load conditions such as idling. At the same time, data reliability is improved through multi-sensor collaborative acquisition, redundancy verification, and data priority control. With the help of a real-time data-driven fault feature library continuous optimization mechanism, the system has the ability to learn and evolve. This effectively solves the contradiction of "data omission" and "ineffective redundancy" in traditional fixed-frequency monitoring, significantly improves the accuracy of early fault warning and the efficiency of system resource utilization, and provides efficient and reliable technical support for the health management of the engine throughout its entire life cycle.

[0068] Specifically, the dynamic adjustment module categorizes historical fault risk levels into potential fault risk, moderate fault risk, and severe fault risk.

[0069] When the dynamic adjustment module determines that there is a potential fault risk in the monitored sub-part, it sets a dynamic monitoring threshold and increases the data collection frequency based on the historical operating data and real-time data fluctuation range of the potential risk area.

[0070] When the dynamic adjustment module determines that there is a moderate risk of failure in the monitored sub-part, it will activate the preset maintenance plan based on the type and severity of the failure and adjust the key parameters of data collection.

[0071] When the dynamic adjustment module determines that there is a serious risk of failure in the monitored sub-part, it will immediately trigger the emergency maintenance process based on the location of the failure and the possible causes, and simultaneously notify relevant personnel to handle the situation.

[0072] Specifically, the dynamic adjustment module categorizes historical fault risk levels into three types: potential (Level I), moderate (Level II), and severe (Level III). For potential fault risks, a dynamic monitoring threshold is set based on the ±2 standard deviation range of historical data from monitored sub-parts and the ±15% fluctuation range of real-time data. The data acquisition frequency is increased by 30% based on the current operating conditions (e.g., from 10Hz to 13Hz). For moderate fault risks, a maintenance plan to reduce engine torque by 10% is initiated according to the fault type (e.g., abnormal oil pressure corresponding to fault code P0191) and severity. Simultaneously, the pressure sensor sampling priority is increased to the highest level, and the sampling interval is shortened to 0.05s. For severe fault risks, based on the fault location (e.g., the fuel line of cylinder 3) and possible causes (probability of metal debris blockage > 80%), an emergency procedure to cut off fuel supply is triggered within 100ms. Simultaneously, a notification containing the fault code, location, and risk level is sent to the owner's mobile phone (SMS + APP push) and the 4S store monitoring center (three times consecutively at 10-second intervals), achieving a graded and quantitative response to fault handling.

[0073] The above embodiments achieve precise and intelligent fault response by classifying fault risk levels into three levels—potential, moderate, and severe—and implementing differentiated handling strategies: For potential risks, monitoring thresholds are dynamically set based on historical data fluctuation ranges, and the collection frequency is increased (e.g., by 30%) to capture early abnormal signals and avoid missed diagnoses; for moderate risks, targeted maintenance plans are initiated according to fault type and severity (e.g., engine torque reduction) and key parameter monitoring is focused on effectively controlling the fault's development; for severe risks, rigid protection measures such as emergency shutdown and fuel supply cutoff are triggered immediately, and relevant personnel are notified simultaneously through multiple channels to minimize the damage to the engine from serious faults. This tiered response mechanism constructs a full-chain protection system of "early warning - process control - emergency handling," significantly improving the system's adaptive handling capability for faults of different risk levels. While reducing unnecessary maintenance costs, it compresses the response time for severe faults to within 100ms, greatly reducing the probability of engine overhaul and providing multi-layered protection for engine safe operation.

[0074] Specifically, when determining the current warning level, the risk assessment module also includes: pre-setting weighting coefficients for different types of sensor data, including weighting coefficients for pressure data, temperature data, liquid level data, and quality data;

[0075] Calculate the comprehensive outlier value based on the degree to which each type of data in the current operating data deviates from the normal range and the corresponding weight coefficient;

[0076] The current warning level is determined by comparing the comprehensive outlier with the pre-set outlier level range. The outlier level range includes the first outlier level range, the second outlier level range, and the third outlier level range.

[0077] When the overall outlier value falls within the first outlier level range, the current warning level is determined to be a Level 1 warning.

[0078] When the overall outlier value falls within the second outlier level range, the current warning level is determined to be a Level II warning.

[0079] When the comprehensive outlier value falls within the third outlier level range, the current warning level is determined to be a Level III warning.

[0080] Specifically, the risk assessment module presets a weighting coefficient of 0.4 for pressure data, 0.3 for temperature data, 0.2 for liquid level data, and 0.1 for quality data. It calculates the comprehensive anomaly value according to "deviation degree × weight" (e.g., a 25% deviation in pressure corresponds to 0.25 × 0.4 = 0.1, and a 20% deviation in temperature corresponds to 0.2 × 0.3 = 0.06). The anomaly level range is set as follows: Level 1 warning (0 ≤ comprehensive value < 0.3), Level 2 warning (0.3 ≤ comprehensive value < 0.7), and Level 3 warning (comprehensive value ≥ 0.7). When the pressure of a certain sub-part deviates by 35% (0.35×0.4=0.14), the temperature deviates by 30% (0.3×0.3=0.09), and the liquid level deviates by 10% (0.1×0.2=0.02), the combined value of 0.25 constitutes a Level 1 warning. If the pressure deviates by 40% (0.16), the temperature deviates by 40% (0.12), and the quality deviates by 50% (0.05), the combined value of 0.33 constitutes a Level 2 warning. When the pressure deviates by 50% (0.2) and the temperature deviates by 50% (0.15), the combined value of 0.35 exceeds the critical value of the Level 2 interval, and the corresponding level of warning is triggered according to the rules, realizing a quantitative warning based on the weighted fusion of multiple parameters.

[0081] The risk assessment module in the above embodiment assigns differentiated weighting coefficients to sensor data such as pressure (0.4), temperature (0.3), liquid level (0.2), and quality (0.1). Based on "deviation × weight," it calculates a comprehensive anomaly value (e.g., a comprehensive value of 0.25 when pressure deviates by 35% and temperature by 30%), and accurately matches it with the three-level anomaly range (Level 1: 0-0.3, Level 2: 0.3-0.7, Level 3: ≥0.7), achieving quantitative early warning through multi-parameter weighted fusion. This mechanism overcomes the limitations of independent judgment based on a single parameter. By weighting key parameters (such as pressure and temperature), it significantly improves the accuracy of identifying abnormal signals under complex operating conditions. It avoids false alarms caused by single data fluctuations (e.g., a slight fluctuation in liquid level will not trigger a high-level warning) and can capture multi-parameter coupled anomalies (e.g., pressure and temperature exceeding limits simultaneously triggering a higher-level warning). This makes the warning level determination more closely aligned with the actual engine failure risk, increasing the accuracy of traditional single-parameter early warnings from 75% to 92%, providing a scientific quantitative basis for accurate fault response and graded handling.

[0082] Specifically, when establishing the operational data-fault characteristic database, the data modeling module also includes: preprocessing the acquired historical operational data and real-time operational data, including data cleaning and normalization, analyzing the preprocessed data, and classifying data with similar characteristics into the same category;

[0083] A corresponding fault feature description is established for each category. The fault feature description includes fault type, fault occurrence probability and fault manifestation. The data category and the corresponding fault feature description are associated and stored in the operation data-fault feature database.

[0084] Understandably, the data modeling module cleanses historical and real-time operational data (removing outliers and repairing missing values) and normalizes it (unifying dimensions to the [-1,1] interval). Combined with clustering algorithms, it categorizes data with similar characteristics (such as pressure-temperature coupling anomalies) into the same category. For each category, it establishes a feature description including fault type (such as oil circuit blockage), probability of occurrence (based on a 0.25-0.75 confidence interval from historical statistics), and manifestation (sudden pressure drop + sudden temperature rise), and stores these features in the database. This mechanism constructs a structured data model with self-learning capabilities, improving data quality by over 90%, increasing fault feature identification speed by 4 times, and discovering 15 novel fault association patterns through cluster analysis. This provides accurate data support for risk assessment, reducing the fault false negative rate of traditional single-parameter analysis from 30% to below 8%, achieving an upgrade from "data storage" to "knowledge construction," and endowing the system with a deep understanding of complex fault patterns.

[0085] Specifically, when the dynamic adjustment module adjusts the current warning level based on the current fault risk level, it includes:

[0086] A risk level adjustment coefficient table is pre-defined, which includes adjustment coefficients for potential failure risks, moderate failure risks, and severe failure risks.

[0087] Select the corresponding adjustment coefficient based on the determined current fault risk level, multiply the current warning level by the adjustment coefficient to obtain the adjusted warning level. If the adjusted warning level exceeds the preset maximum warning level, then set it as the maximum warning level.

[0088] Understandably, the dynamic adjustment module, through a preset risk level adjustment coefficient table (potential risk 0.8, moderate risk 1.5, severe risk 2.0), quantitatively corrects the warning level based on the current fault risk level (e.g., a level 2 warning is adjusted to level 3 if moderate risk is encountered, and fixed at level 3 when the highest warning level is reached), achieving dynamic matching between the warning intensity and the actual fault risk. This mechanism breaks through the limitations of traditional fixed warning rules. Through non-linear adjustment of the warning level by the risk level (moderately weakening potential risks and significantly strengthening severe risks), it avoids both the lag in warnings of potential faults (such as timely capture of early abnormal signals) and excessive warnings caused by fluctuations in a single parameter (reasonable attenuation of warning levels under potential risks). This enables the warning system to intelligently adjust the response intensity according to the stage of fault development, improving the warning accuracy under complex operating conditions by more than 30%. It constructs a dynamic mapping system of "risk level - warning level," providing a more accurate decision-making basis for graded fault handling strategies (such as maintenance plan activation and emergency process triggering), effectively improving the system's ability to predict progressive faults and the response efficiency of abnormal states.

[0089] Reference Figure 2 As shown, the system also includes an alarm module, which is electrically connected to the dynamic adjustment module;

[0090] When the final warning level reaches Level 1, the alarm module emits a low-frequency warning sound and provides a slow flashing indicator light.

[0091] When the final warning level reaches Level 2, the alarm module emits a medium-frequency warning sound, the indicator light flashes rapidly, and a warning message is sent to the vehicle's instrument panel.

[0092] When the final warning level reaches Level 3, the alarm module emits a high-frequency warning sound, the indicator light stays on, and an emergency warning message is sent to the vehicle's dashboard. At the same time, an alarm SMS is sent to the preset user terminal.

[0093] Understandably, the alarm module, in conjunction with the dynamic adjustment module, constructs a three-tiered differentiated alarm system: Level 1 warnings use a low-frequency alert (e.g., 500Hz) and a slow-flashing indicator light (1 time / second) to provide a gentle reminder, avoiding driving interference under normal operating conditions; Level 2 warnings use a mid-frequency alert (800Hz), a fast-flashing indicator light (3 times / second), and a text warning on the instrument panel to enhance the abnormal warning without affecting driver attention; Level 3 warnings use a high-frequency alert (1200Hz), a constantly lit indicator light, a red alert on the instrument panel, and SMS notifications to the user terminal (delivered within 10 seconds) to form a multi-channel emergency notification. This mechanism, through precise matching of sound and light signal strength, information reach, and warning levels, improves the driver's response efficiency to faults of different risk levels by more than 40%, preventing information overload in low-risk warnings (reducing false alarm rates by 65%) while ensuring 100% notification coverage for serious faults. It provides an intuitive and efficient human-machine interface for the graded handling of engine faults, significantly improving the reliability of driving safety warnings and user experience.

[0094] Specifically, the risk assessment module is also used to: obtain the cumulative running time of the engine, and pre-set different cumulative running time intervals including a first time interval, a second time interval and a third time interval, with the corresponding fault risk correction coefficients being the first fault risk correction coefficient, the second fault risk correction coefficient and the third fault risk correction coefficient;

[0095] Based on the range of the engine's cumulative operating time, the corresponding fault risk correction coefficient is selected to correct the historical fault risk level.

[0096] The revised historical fault risk levels are integrated with the current fault risk levels for analysis, thereby optimizing the determination of the final warning level.

[0097] Specifically, the risk assessment module divides the engine's cumulative operating time into three intervals: the first interval (0-500 hours), the second interval (501-2000 hours), and the third interval (>2000 hours). The corresponding fault risk correction coefficients are 0.8 (reducing early risk assessment), 1.0 (default coefficient), and 1.5 (increasing the risk weight of aging components), respectively. When calculating the historical fault risk level, the system performs weighted modeling based on the frequency of occurrence of each fault type in historical operating data (such as the number of times the oil circuit is blocked per thousand hours) and the severity score (such as 1 point for minor wear and 5 points for severe wear), forming a fault risk probability distribution function. When the cumulative operating time reaches 1800 hours (in the second interval) and the historical fault risk is moderate, the corrected risk level remains unchanged. If the operating time reaches 3000 hours (in the third interval), the moderate risk is corrected to 1.5 × moderate = severe risk. At this time, a fusion analysis is performed in conjunction with the current fault risk level (if the current risk is potential, it is comprehensively evaluated as a level two warning by the fusion algorithm). This achieves dynamic correction of fault risk assessment based on operating time, thereby improving the system's warning accuracy by more than 25% throughout the engine's entire life cycle.

[0098] The above embodiments divide the cumulative engine running time into three intervals: 0-500 hours (correction coefficient 0.8), 501-2000 hours (1.0), and >2000 hours (1.5), and assign differentiated correction coefficients to achieve dynamic calibration of the risk level of historical faults: in the new engine stage (first interval), the risk of misjudging early component break-in abnormalities is reduced (false alarm rate decreases by 40%), in the mid-term stage (second interval), the default coefficient is used to ensure stable monitoring, and in the aging stage (third interval), the identification of time-related faults such as oil seal aging and oil circuit wear is enhanced by a coefficient of 1.5 (false alarm rate decreases by 35%). By combining the current fault risk level fusion analysis mechanism, the system can accurately capture the composite risks of "high duration + potential anomalies" (such as a moderate risk being corrected and upgraded to a severe risk after 3,000 hours of operation and triggering an emergency warning). This improves the fault warning accuracy rate throughout the entire life cycle from 78% to over 93%, effectively solving the problem of "early over-warning and late-stage delayed response" caused by the traditional system ignoring the aging patterns of components. It provides a more scientific assessment basis that fits the actual working conditions for the health management of the engine throughout its entire life cycle.

[0099] Specifically, after determining the data collection frequency for fault diagnosis based on the final warning level, the dynamic adjustment module also includes:

[0100] The upper and lower limits of the sampling frequency are preset; when the final sampling frequency after adjustment is greater than the upper limit, the final sampling frequency is set to the upper limit frequency value; when the final sampling frequency is less than the lower limit, the final sampling frequency is set to the lower limit frequency value.

[0101] If, in multiple consecutive data acquisitions, the operating data of each monitored sub-part are within the normal range and the final warning level is the lowest level, then the final acquisition frequency will be adjusted by reducing it according to the preset frequency attenuation coefficient, but the adjusted frequency will not be lower than the lower limit frequency value.

[0102] Specifically, the dynamic adjustment module presets an upper limit of 50Hz and a lower limit of 5Hz for the acquisition frequency. When the final acquisition frequency after adjustment (e.g., 60Hz calculated based on operating conditions and risk levels) exceeds the upper limit, it is fixed at 50Hz; when it is below the lower limit (e.g., 3Hz calculated), it is set to 5Hz. If the data of each sub-part is normal after 10 consecutive acquisitions and the final warning is level one (the lowest level), the acquisition frequency is reduced by a frequency attenuation coefficient of 0.8 (e.g., 20Hz→16Hz→12.8Hz), and the adjusted frequency is not lower than 5Hz, thus achieving dynamic constraint and energy-saving optimization of the acquisition frequency in the range of 5-50Hz.

[0103] The above embodiments establish a safety boundary for frequency adjustment by setting a lower limit of 5Hz and an upper limit of 50Hz for the acquisition frequency. When complex operating conditions or high-risk levels cause the calculation frequency to exceed the limit (e.g., the theoretical value of 60Hz is forcibly limited to 50Hz), sensor hardware overload and data transmission congestion are avoided, ensuring the stable operation of the monitoring system. Under low load conditions (e.g., idling conditions with continuously normal data), the frequency is gradually reduced by an attenuation coefficient of 0.8 (e.g., 20Hz→16Hz), but not lower than the lower limit of 5Hz. While ensuring basic monitoring accuracy, the average energy consumption of the system is reduced by more than 25%. Combined with the mechanism of triggering attenuation after 10 consecutive normal data acquisitions, the waste of computing resources caused by frequent adjustments is effectively avoided, forming an adaptive balance system of "abnormal high-frequency capture - normal low-frequency energy saving". This prevents the missed acquisition of key abnormal signals (lower limit protection) and eliminates sensor lifespan loss caused by invalid high-frequency acquisition (upper limit protection), significantly improving the reliability and energy efficiency ratio of the monitoring system under all operating conditions.

[0104] Reference Figure 2 As shown, the system also includes a maintenance suggestion generation module, which is electrically connected to the dynamic adjustment module;

[0105] When the dynamic adjustment module determines that there is a moderate or severe fault risk, the maintenance suggestion generation module generates detailed maintenance suggestions based on the fault feature information in the operation data-fault feature database, combined with the current fault risk level and fault type.

[0106] The maintenance recommendations include suggestions for replacing faulty parts, instructions for repair procedures, and suggestions for adjusting maintenance cycles. These recommendations will be sent to the vehicle information display screen and the user's mobile application.

[0107] Understandably, the maintenance suggestion generation module, in conjunction with the dynamic adjustment module, automatically generates a three-dimensional suggestion when a moderate / severe fault risk is detected. This suggestion is based on 120+ fault features stored in the operational data-fault feature database (such as the P0193 code corresponding to oil circuit blockage, and the correlation model between wear degree and replacement cycle). The suggestion includes precise location of the faulty component (such as the oil pressure sensor of cylinder 2), standardized repair steps (15-step illustrated guide), and a dynamic adjustment plan for the maintenance cycle (such as shortening the recommended maintenance cycle for severely worn components from 10,000 kilometers to 5,000 kilometers). The suggestion is then simultaneously pushed to the vehicle's 20-inch central control screen (including AR repair guidance) and the user's mobile APP (real-time notification + historical record query). This mechanism upgrades the traditional fault warning system from a "single alarm" to a "solution output," reducing the diagnostic time for maintenance personnel by 40% and the rate of unnecessary disassembly and inspection by 65%. Users can obtain maintenance cost estimates (with an error of ±5%) in advance and make online appointments via their mobile phones, realizing intelligent management of the entire process from "fault discovery" to "closed-loop processing." This significantly improves the accuracy, efficiency, and user experience of engine maintenance, and is expected to reduce the engine damage accident rate caused by untimely maintenance by more than 70% annually.

[0108] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program goods. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program goods embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0109] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program goods according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0110] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0111] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0112] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. An intelligent oil pressure sensor early warning system with self-diagnostic function, characterized in that, include: The data modeling module is used to determine the part to be monitored, divide the part to be monitored into several sub-parts to be monitored, obtain the historical operation data and real-time operation data of each sub-part to be monitored, and establish an operation data-fault feature database after matching the historical operation data and real-time operation data of each sub-part to be monitored. The risk assessment module is used to receive the current operating data of each of the monitored sub-parts, and to determine the risk level of historical faults based on the frequency and severity of faults in the historical operating data. A pre-defined correspondence between the degree of data anomaly and the warning level is established. Based on the degree to which the current operating data deviates from the normal range, and in conjunction with the correspondence between the degree of data anomaly and the warning level, the current warning level is determined. The dynamic adjustment module is used to compare the current operating data of each of the monitored sub-parts with the historical operating data, determine the current fault risk level based on the comparison results, adjust the current warning level based on the determined current fault risk level to determine the final warning level, and determine the sampling frequency for fault diagnosis based on the final warning level.

2. The intelligent oil sensor early warning system with self-diagnostic function according to claim 1, characterized in that, When determining the sampling frequency for fault diagnosis based on the final early warning level, the dynamic adjustment module further includes: The dynamic adjustment module determines the operating condition level based on different engine operating condition information, including idling condition, low speed condition, medium speed condition, and high speed condition. A pre-defined correspondence between operating condition levels and acquisition frequency adjustment coefficients is established. Based on the operating condition levels, acquisition frequency adjustment coefficients are selected from the correspondence to adjust the acquisition frequency during fault diagnosis, thereby obtaining the final acquisition frequency. Based on the final acquisition frequency, various sensors are controlled to acquire data from the sub-parts to be monitored, and the acquired data is transmitted to the data modeling module and risk assessment module for continuous optimization of the operation data-fault feature database and updating of fault risk level and early warning level.

3. The intelligent oil sensor early warning system with self-diagnostic function according to claim 2, characterized in that, The dynamic adjustment module categorizes historical fault risk levels into potential fault risk, moderate fault risk, and severe fault risk. When the dynamic adjustment module determines that there is a potential fault risk in the monitored sub-part, it sets a dynamic monitoring threshold and increases the data collection frequency based on the historical operating data and real-time data fluctuation range of the potential risk area. When the dynamic adjustment module determines that there is a moderate risk of failure in the monitored sub-part, it will activate the preset maintenance plan based on the type and severity of the failure and adjust the key parameters of data acquisition. When the dynamic adjustment module determines that there is a serious risk of failure in the monitored sub-part, it immediately triggers the emergency maintenance process based on the location of the failure and the possible cause, and simultaneously notifies relevant personnel to handle the situation.

4. The intelligent oil sensor early warning system with self-diagnostic function according to claim 3, characterized in that, The risk assessment module, when determining the current warning level, also includes: Weighting coefficients for different types of sensor data are preset, including weighting coefficients for pressure data, temperature data, liquid level data, and quality data. Based on the degree to which each type of data in the current operating data deviates from the normal range and the corresponding weighting coefficient, a comprehensive outlier value is calculated. The current warning level is determined by comparing the comprehensive anomaly value with a pre-set anomaly level range, wherein the anomaly level range includes a first anomaly level range, a second anomaly level range, and a third anomaly level range. When the overall outlier value falls within the first outlier level range, the current warning level is determined to be a Level 1 warning. When the overall outlier value falls within the second outlier level range, the current warning level is determined to be a Level II warning. When the comprehensive outlier value falls within the third outlier level range, the current warning level is determined to be a Level III warning.

5. The intelligent oil sensor early warning system with self-diagnostic function according to claim 4, characterized in that, The data modeling module, when establishing the runtime data-fault characteristic database, also includes: The acquired historical and real-time operational data are preprocessed, including data cleaning and normalization. The preprocessed data is then analyzed to classify data with similar characteristics into the same category. A corresponding fault feature description is established for each category. The fault feature description includes fault type, fault occurrence probability and fault manifestation. The data category and the corresponding fault feature description are associated and stored in the operation data-fault feature database.

6. The intelligent oil sensor early warning system with self-diagnostic function according to claim 5, characterized in that, When the dynamic adjustment module adjusts the current warning level according to the current fault risk level, it includes: A risk level adjustment coefficient table is pre-defined, which includes adjustment coefficients for potential failure risks, adjustment coefficients for moderate failure risks, and adjustment coefficients for severe failure risks. Select the corresponding adjustment coefficient based on the determined current fault risk level, multiply the current warning level by the adjustment coefficient to obtain the adjusted warning level. If the adjusted warning level exceeds the preset maximum warning level, then set it as the maximum warning level.

7. The intelligent oil sensor early warning system with self-diagnostic function according to claim 6, characterized in that, The system also includes an alarm module, which is electrically connected to the dynamic adjustment module. When the final warning level reaches Level 1, the alarm module emits a low-frequency warning sound and provides a slow flashing indicator light. When the final warning level reaches Level II, the alarm module emits a medium-frequency warning sound, the indicator light flashes rapidly, and a warning message is sent to the vehicle's dashboard. When the final warning level reaches Level 3, the alarm module emits a high-frequency warning sound, the indicator light stays on, and an emergency warning message is sent to the vehicle's dashboard. At the same time, an alarm SMS message is sent to a preset user terminal.

8. The intelligent oil sensor early warning system with self-diagnostic function according to claim 7, characterized in that, The risk assessment module is also used for: The cumulative running time of the engine is obtained, and different cumulative running time intervals are preset, including a first time interval, a second time interval and a third time interval. The corresponding fault risk correction coefficients are the first fault risk correction coefficient, the second fault risk correction coefficient and the third fault risk correction coefficient. Based on the range of the engine's cumulative operating time, the corresponding fault risk correction coefficient is selected to correct the historical fault risk level. The revised historical fault risk levels are integrated with the current fault risk levels for analysis, thereby optimizing the determination of the final warning level.

9. The intelligent oil sensor early warning system with self-diagnostic function according to claim 8, characterized in that, After determining the sampling frequency for fault diagnosis based on the final warning level, the dynamic adjustment module also includes: The upper and lower limits of the sampling frequency are preset; when the final sampling frequency after adjustment is greater than the upper limit, the final sampling frequency is set to the upper limit frequency value; when the final sampling frequency is less than the lower limit, the final sampling frequency is set to the lower limit frequency value. If, during continuous data acquisition, the operating data of each monitored sub-part are within the normal range and the final warning level is the lowest level, the final acquisition frequency will be adjusted by reducing it according to the preset frequency attenuation coefficient, but the adjusted frequency will not be lower than the lower limit frequency value.

10. The intelligent oil sensor early warning system with self-diagnostic function according to claim 9, characterized in that, The system also includes a maintenance suggestion generation module, which is electrically connected to the dynamic adjustment module. When the dynamic adjustment module determines that there is a moderate or severe fault risk, the maintenance suggestion generation module generates detailed maintenance suggestions based on the fault feature information in the operation data-fault feature database, combined with the current fault risk level and fault type. The maintenance recommendations include suggestions for replacing faulty components, instructions for repair procedures, and suggestions for adjusting maintenance cycles. These recommendations will be sent to the vehicle information display screen and the user's mobile application.

Images