An engine real-time multi-working condition intelligent detection intuitive figure drawing method and system

Abstract

The application discloses a kind of engine real-time multi-condition intelligent detection intuitive figure drawing method and system, it is related to data analysis technical field.The method includes: the multidimensional real-time running data stream of target engine is collected, characteristic extraction calibration is executed to multidimensional real-time running data stream, and engine operation multidimensional feature vector is generated;Carry out condition boundary self-adapting division, and construct engine dynamic multi-condition detector;Real-time condition detection is carried out to engine operation multidimensional feature vector, and current timing engine condition is determined;Introduce abnormal figure enhancement mechanism, and carry out figure correlation mapping and layered drawing enhancement in combination with engine universal characteristic curve and current timing engine condition, and generate engine condition layered visualized figure.The technical problems that engine condition detection in prior art is difficult to adapt to real-time dynamic change, and the visual degree of detection result is lower are solved, the technical effects that the accuracy of engine multi-condition detection and operation state intuitive display capability are improved are achieved.

Description

Technical Field

[0001] This invention relates to the field of data analysis technology, specifically to a method and system for intuitive graphical rendering of real-time multi-condition intelligent detection of engines. Background Technology

[0002] With the intelligent development of automotive powertrain systems, parameters such as engine speed, load, temperature, and pressure exhibit significant dynamic changes during engine operation. Existing engine condition monitoring methods typically rely on preset operating ranges or fixed parameter thresholds for judgment, primarily depending on offline data analysis or empirical rules for operating condition classification. This makes it difficult to accurately reflect the dynamic transitions between various operating conditions during actual vehicle operation. Furthermore, existing detection results are mostly presented as parameter lists or simple curves, lacking a clear correlation with engine performance characteristics. This results in a less intuitive operational status analysis process and makes it difficult to promptly detect abnormal operating trends. Summary of the Invention

[0003] This application provides a method and system for intuitive graphical rendering of real-time multi-condition intelligent detection of engines, which solves the technical problems in the prior art where engine condition detection is difficult to adapt to real-time dynamic changes and the visualization of detection results is low.

[0004] The first aspect of this application provides a method for intuitive graphical representation of real-time multi-condition intelligent detection of an engine, the method comprising: A multi-dimensional real-time operating data stream of the target engine is collected, and feature extraction and calibration are performed on the multi-dimensional real-time operating data stream to generate a multi-dimensional feature vector of engine operation. An engine dynamic multi-condition detector is constructed by adaptively dividing the operating condition boundary by combining the engine's universal characteristic curve and the vehicle's actual operating characteristic data. The engine dynamic multi-condition detector is used to perform real-time operating condition detection on the engine's multi-dimensional feature vector to determine the current time-series engine operating condition. An abnormal graphics enhancement mechanism is introduced to combine the engine's universal characteristic curve and the current time-series engine operating condition for graphics association mapping and layered rendering enhancement, generating a layered visualization graphics of the engine operating condition.

[0005] A second aspect of this application provides an intuitive graphical rendering system for real-time multi-condition intelligent detection of engines, the system comprising: Data calibration module: Collects multi-dimensional real-time operating data streams of the target engine, performs feature extraction and calibration on the multi-dimensional real-time operating data streams, and generates multi-dimensional feature vectors of engine operation; Boundary division module: Combines the engine universal characteristic curve and the actual operating characteristic data of the vehicle to perform adaptive division of operating condition boundaries and constructs an engine dynamic multi-condition detector; Operating condition detection module: Uses the engine dynamic multi-condition detector to perform real-time operating condition detection on the engine operating multi-dimensional feature vectors to determine the current time-series engine operating condition; Visualization module: Introduces an abnormal graphics enhancement mechanism to combine the engine universal characteristic curve and the current time-series engine operating condition to perform graphic association mapping and layered drawing enhancement, generating layered visualization graphics of engine operating conditions.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: First, a multi-dimensional real-time operating data stream of the target engine is collected, and feature extraction and calibration are performed on the multi-dimensional real-time operating data stream to generate a multi-dimensional feature vector of engine operation. Next, by combining the engine's universal characteristic curve and the actual operating characteristic data of the vehicle, adaptive boundary delineation of operating conditions is performed to construct a dynamic multi-condition engine detector. Then, the dynamic multi-condition engine detector is used to perform real-time operating condition detection on the multi-dimensional feature vector of engine operation to determine the current time-series engine operating condition. Finally, an anomaly graph enhancement mechanism is introduced to combine the engine's universal characteristic curve and the current time-series engine operating condition for graph association mapping and layered rendering enhancement, generating a layered visualization graph of engine operating conditions. This solves the technical problems of existing technologies where engine operating condition detection is difficult to adapt to real-time dynamic changes and the visualization of detection results is low, achieving the technical effect of improving the accuracy of multi-condition engine detection and the ability to intuitively display the operating status. Attached Figure Description

[0007] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0008] Figure 1 This application provides a schematic flowchart of an intuitive graphical drawing method for real-time multi-condition intelligent detection of an engine. Figure 2 A schematic diagram of the universal characteristic curves of an engine provided in the embodiments of this application; Figure 3 This is a schematic diagram of the structure of an intuitive graphical drawing system for real-time multi-condition intelligent detection of an engine, provided in an embodiment of this application.

[0009] Figure labeling: Data calibration module 11, boundary division module 12, working condition detection module 13, visualization module 14. Detailed Implementation

[0010] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0011] Example 1, as Figure 1 As shown, this application provides a method for intuitive graphical rendering of real-time multi-condition intelligent detection of an engine, wherein the method includes: Collect multidimensional real-time operating data streams of the target engine, perform feature extraction and calibration on the multidimensional real-time operating data streams, and generate multidimensional feature vectors of engine operation.

[0012] Through the data acquisition interface deployed in the target engine and vehicle control system, multi-dimensional operational data streams during engine operation are acquired in real time. These multi-dimensional real-time operational data streams include at least engine speed data, engine output torque data, fuel injection quantity data, intake air pressure data, intake air temperature data, coolant temperature data, throttle opening data, and vehicle operating load data. Engine speed data and output torque data are used to characterize the current operating power state of the engine, fuel injection quantity data and intake air parameters are used to characterize the engine combustion control state, temperature parameters are used to characterize the engine thermal operating state, and vehicle operating load data are used to characterize the factors affecting the overall vehicle operating condition.

[0013] Furthermore, feature extraction and calibration are performed on the multidimensional real-time operational data stream to generate a multidimensional feature vector of engine operation, including: Based on the data source types of the multidimensional real-time operating data stream, a preprocessing step is performed to obtain the multidimensional data preprocessing step process. The data source types include steady-state operating data, transient change data, and vehicle linkage data. Hierarchical preprocessing is performed on the multidimensional real-time operating data stream according to the multidimensional data preprocessing step process to obtain a standard multidimensional operating data stream. An engine multidimensional correlation feature set is extracted from the standard multidimensional operating data stream. Normalization and time-series calibration are performed on the engine multidimensional correlation feature set to generate an engine operating multidimensional feature vector.

[0014] Preferably, a multi-dimensional real-time operational data stream is acquired during the operation of the target engine. This multi-dimensional real-time operational data stream includes engine speed n, output torque T, and fuel injection data collected by the engine electronic control unit. Intake pressure Intake air temperature Coolant temperature Throttle opening The system also collects operating parameters such as vehicle speed (v) and road load (L), and performs preprocessing steps based on the data source type of each operating parameter. The multidimensional real-time operating data stream is then divided into steady-state operating data, transient change data, and vehicle-wide linkage data. For steady-state operating data, a fixed time window is used to calculate the mean and volatility of continuously sampled data to obtain the steady-state characteristics of the corresponding parameters, such as the mean engine speed. Average torque and parameter standard deviation This is used to characterize the stable operating range of the engine; for transient data, dynamic change characteristics are calculated based on the data difference between adjacent sampling times, such as the rate of change of engine speed. ,in, Indicates the rate of change of engine speed; This indicates the engine speed at the current sampling time; This represents the engine speed at the previous sampling time. This indicates the sampling time interval; simultaneously, engine load response characteristics are calculated based on torque changes and fuel injection quantity changes to describe the engine operating condition transition process; for vehicle linkage data, parameters such as vehicle speed, road load, and throttle opening are synchronously matched according to timestamps to calculate the correlation characteristics between vehicle operating state and engine output state.

[0015] Based on the data processing requirements corresponding to the different types of data mentioned above, a multidimensional data preprocessing process is generated, and hierarchical preprocessing is performed according to the multidimensional data preprocessing process. First, time alignment is performed on data with different sampling frequencies using a timestamp calibration method, mapping each data to a unified sampling sequence. Then, an anomaly detection method based on a sliding window is used to calculate the mean and deviation of parameters within the window. When a sampled value deviates from a preset deviation range, it is marked as an anomaly, and correction is performed using interpolation of adjacent valid sampling points. Simultaneously, low-pass filtering is applied to sensor data with high-frequency disturbances to obtain a continuous, synchronous, and noise-free standard multidimensional operating data stream. After obtaining the standard multidimensional operating data stream, correlation features are extracted from the engine operating parameters, and engine output power features are calculated based on speed and torque. ,in, Indicates engine output power; This indicates the engine's output torque; This represents engine speed; 9550 is the unit conversion factor; simultaneously, fuel consumption characteristics are calculated based on fuel injection quantity and output power, and thermal state correction characteristics are calculated based on intake pressure, intake temperature, and coolant temperature. Steady-state characteristics, transient response characteristics, and vehicle linkage characteristics are combined to form a multi-dimensional engine correlation feature set. Furthermore, the feature parameters in the multi-dimensional engine correlation feature set are normalized, converting feature parameters of different dimensions to a unified value range, for example, using a maximum-minimum normalization method. ,in, Represents the normalized i-th eigenvalues; Represents the original eigenvalues; This indicates the maximum value of the feature; This represents the minimum value of the feature; subsequently, time-series calibration is performed based on the acquisition timestamps corresponding to each feature parameter, establishing a correspondence between each feature value and the real-time operation time of the engine, ultimately generating a multi-dimensional feature vector of engine operation: ,in, Indicates time The corresponding multidimensional feature vector of engine operation; This represents the engine operating characteristic parameter m after normalization and time-series calibration. Through the above processing, the multi-dimensional feature vector of engine operation can simultaneously contain the engine's steady-state operating state, transient response changes, and the impact information on vehicle operation, providing a data foundation for subsequent dynamic operating condition boundary division and real-time operating condition detection based on the engine's universal characteristic curve.

[0016] By combining the universal characteristic curve of the engine and the actual operating characteristic data of the vehicle, the working condition boundary is adaptively divided, and a dynamic multi-working condition detector for the engine is constructed.

[0017] Furthermore, by combining the engine's universal characteristic curves and actual vehicle operating characteristic data, adaptive boundary delineation of operating conditions is performed to construct a dynamic multi-condition engine detector, including: Import the engine universal characteristic curve, align the coordinate system of the engine universal characteristic curve with speed as the horizontal axis and torque as the vertical axis to obtain the usable engine universal characteristic curve; anchor reference feature points, layered contour lines of fuel consumption, and full-load operating condition boundary contours on the usable engine universal characteristic curve, wherein the reference feature points include the idle speed operating point, the maximum torque operating point, and the rated power operating point; based on the reference feature points, layered contour lines of fuel consumption, and full-load operating condition boundary contours, generate a universal characteristic reference base map; project the actual vehicle operating characteristic data onto the universal characteristic reference base map to perform adaptive division of operating condition boundaries, and construct an engine dynamic multi-operating condition detector.

[0018] Preferably, an engine feature identifier is established based on the engine model information, displacement parameters, rated power parameters, maximum torque parameters, and fuel type parameters of the target engine. Engine test calibration data corresponding to the engine feature identifier is retrieved from a preset engine characteristic database. This engine test calibration data includes output torque data, fuel consumption rate data, and load state data under different engine speed conditions. A three-dimensional correlation dataset of engine speed-torque-fuel consumption rate is established based on the engine test calibration data. Using engine speed as the horizontal coordinate and output torque as the vertical coordinate, the fuel consumption rate data corresponding to different speed and torque combinations is mapped to a two-dimensional coordinate region, generating the engine universal characteristic curve data corresponding to the target engine. Further, the engine universal characteristic curve data undergoes coordinate system alignment processing. The coordinate mapping ratio is determined based on the speed range and torque range of the actual operating data of the target engine. The speed coordinates and torque coordinates in the engine universal characteristic curve are scaled to ensure that the engine universal characteristic curve and the actual operating characteristic data of the vehicle are in the same speed-torque coordinate system, resulting in a usable engine universal characteristic curve that can be used for operating condition analysis.

[0019] On the available universal characteristic curves of the engine, reference feature points are determined based on key operating states in the engine test calibration data. The speed and torque positions corresponding to the lowest stable operating state of the engine are determined as the idle speed condition point, the maximum torque output position is determined as the maximum torque condition point, and the rated power output position is determined as the rated power condition point. At the same time, based on the speed-torque combination data corresponding to different fuel consumption rates during the engine test, multi-layer equal-height fuel consumption lines are generated using an equal-value connection method. The full-load condition boundary contour line is determined based on the engine's maximum fuel supply and maximum load test data. The reference feature points, multi-layer equal-height fuel consumption lines, and full-load condition boundary contour line are superimposed on a unified coordinate space to form a universal characteristic reference base map. Subsequently, the engine speed data and output torque data collected during the actual operation of the vehicle are converted into corresponding coordinate points and projected onto the universal characteristic reference base map. By statistically analyzing the distribution number and spatial clustering of actual operating data points in different coordinate areas, the main operating areas during the actual operation of the vehicle are determined. Based on the deviation relationship between the actual operating area and the original operating condition boundary, the operating condition boundary is dynamically adjusted to obtain an adaptive operating condition boundary that matches the actual operating state of the target engine. Based on the adaptive operating condition boundary, the actual operating data is labeled with operating conditions to construct an engine dynamic multi-operating condition detector.

[0020] like Figure 2The diagram shows a schematic of the universal characteristic curve of the available engine in this embodiment. First, the original universal characteristic data of the target engine is transformed according to a unified coordinate reference, establishing a two-dimensional coordinate space for engine operating characteristics with engine speed as the abscissa and engine output torque as the ordinate. Within this two-dimensional coordinate space, the boundary contour line of the engine's full-load operating condition and the contour lines corresponding to different fuel consumption levels are loaded to characterize the engine's performance changes under different combinations of speed and torque. Simultaneously, benchmark feature points such as the idle speed operating point, maximum torque operating point, and rated power operating point are marked on the available engine universal characteristic curve. These benchmark feature points determine the spatial distribution of typical engine operating states. In subsequent operating condition testing, the available engine universal characteristic curve is used as the operating condition mapping reference to transform the collected actual vehicle operating characteristic data to the corresponding coordinate positions, realizing the correlation analysis between the vehicle's actual operating state and engine performance characteristics, providing basic data support for the dynamic multi-operating condition region division of the engine.

[0021] Furthermore, the actual vehicle operating characteristic data is projected onto the universal characteristic baseline map for adaptive boundary delineation of operating conditions, constructing an engine dynamic multi-condition detector, including: The actual vehicle operating characteristic data is projected onto the universal characteristic baseline map to generate a real vehicle operating density heatmap; the operating condition boundaries are pre-divided by combining the actual vehicle operating characteristic data and the real vehicle operating density heatmap to obtain initial multi-operating condition region boundary thresholds; a sliding calibration window is activated to adaptively calibrate the initial multi-operating condition region boundary thresholds to determine adaptive multi-operating condition region boundary thresholds; based on the adaptive multi-operating condition region boundary thresholds, the actual vehicle operating characteristic data is labeled and classified for training to construct an engine dynamic multi-operating condition detector.

[0022] Preferably, the engine speed data and output torque data are extracted from the actual vehicle operating characteristic data, and projected and mapped according to the coordinate relationship corresponding to the universal characteristic reference base map, converting each set of actual operating data into two-dimensional coordinate points of speed-torque. The operating coordinate point corresponding to the i-th set of actual vehicle operating characteristic data is represented as... ,in, Indicates engine speed. This represents the engine output torque; subsequently, based on the speed range and torque range of the universal characteristic reference map, a two-dimensional grid is established to divide the region. Each projected operating coordinate point is mapped to its corresponding grid cell, and the number of operating data points within each grid area is counted to calculate the data distribution density for each region. ,in, This represents the running data density of the j-th grid region; This represents the number of running data points that fall within the j-th grid region; This represents the area of ​​the j-th grid region. A real-vehicle operating density heatmap is generated based on the data density values ​​corresponding to each grid region, ensuring that different speed-torque regions in the universal characteristic baseline map reflect the high-frequency and low-frequency operating regions during actual vehicle operation. Further, combining the distribution of operating points corresponding to the actual vehicle operating characteristic data and the real-vehicle operating density heatmap, cluster analysis is performed on the actual engine operating regions. Candidate operating condition regions are determined based on the continuity of data density between adjacent regions. Furthermore, combining the pre-marked idle speed operating point, maximum torque operating point, rated power operating point, and full-load operating condition boundary contour line in the universal characteristic baseline map, boundary constraint corrections are applied to the candidate operating condition regions to determine the initial multi-operating condition region boundary thresholds. These initial multi-operating condition region boundary thresholds include the speed range, torque range, and region transition judgment conditions corresponding to each operating condition region. To improve the adaptability of the operating condition division results to changes in actual vehicle operation, a sliding calibration window is further established. Multiple engine operating coordinate points within the window are continuously acquired according to a preset time length, and the data distribution centroid corresponding to the current window is calculated. ,in, This represents the centroid of the data distribution corresponding to the k-th sliding calibration window; This indicates the average engine speed within the window. The average engine torque within the window is represented. The data distribution centroid corresponding to multiple consecutive sliding windows is compared with the initial multi-condition region boundary. When the data distribution centroid is detected to be continuously shifted and the shift exceeds the preset calibration threshold, the boundary correction amount is calculated based on the data shift direction and the shift amplitude, and the initial multi-condition region boundary threshold is dynamically adjusted to obtain the adaptive multi-condition region boundary threshold. Finally, based on the adaptive multi-condition region boundary threshold, the actual vehicle operating feature data is labeled and classified for training. The speed-torque coordinate points in each group of actual vehicle operating feature data are matched with the adaptive multi-condition region boundary threshold. When the speed-torque coordinate point is within the speed range and torque range corresponding to a certain operating condition region, the category identifier corresponding to that operating condition region is used as the operating condition label of the current operating data, thereby establishing a correspondence between the actual vehicle operating feature data and the operating condition category, forming a classification training sample containing a multi-dimensional feature vector of engine operation and an operating condition category label. Based on the classification training sample, the speed feature, torque feature, load feature, fuel injection feature, and dynamic change feature in the multi-dimensional feature vector of engine operation are used as inputs, and the corresponding operating condition category label is used as the output to establish a mapping relationship between each feature parameter and the operating condition category, and operating condition judgment constraint parameters are generated according to the feature distribution range corresponding to each operating condition category. When a new multi-dimensional feature vector of engine operation is input, the operating feature parameters in it are matched with the operating condition judgment constraint parameters corresponding to each operating condition category. The current engine operating condition category is determined according to the matching result, and the real-time operating condition detection result is output, thereby constructing an engine dynamic multi-condition detector.

[0023] Furthermore, the sliding calibration window is activated to adaptively calibrate the initial multi-condition region boundary threshold, determining the adaptive multi-condition region boundary threshold, including: Initiate a sliding calibration window, collect multiple real-vehicle operating data within multiple consecutive sliding calibration windows in real time, and calculate the average speed and average torque coordinates of the multiple real-vehicle operating data; determine the centroid of the data distribution of multiple windows based on the average speed and average torque coordinates; project the centroid of the data distribution of multiple windows onto the universal characteristic reference base map, and mark multiple operating feature anchor points; when a preset calibration condition is triggered, perform adaptive offset calibration on the initial multi-condition region boundary threshold based on the multiple operating feature anchor points, and determine the adaptive multi-condition region boundary threshold.

[0024] A sliding calibration window is initiated based on a preset time length to dynamically collect continuous real-vehicle operation data during actual vehicle operation. Each sliding calibration window contains engine speed and output torque data corresponding to multiple consecutive sampling moments. The average speed and average torque coordinates corresponding to the current window are calculated based on the data collected within the window. The specific calculation method is as follows: , ,in, This represents the average engine speed within the k-th sliding calibration window; This represents the average engine torque within the k-th sliding calibration window; N represents the number of operating data points contained in the current sliding calibration window. This represents the engine speed corresponding to the i-th set of operating data; This represents the engine output torque corresponding to the i-th set of operating data. Based on the average speed and average torque coordinates, the data distribution center corresponding to each sliding calibration window is determined as the centroid of the window data distribution. ,in, This represents the centroid of the data distribution corresponding to the k-th sliding calibration window. Subsequently, the centroids of the window data distribution corresponding to multiple consecutive sliding calibration windows are sequentially projected onto the speed-torque coordinate space of the universal characteristic reference map. Based on the positional relationship between each centroid and the original operating condition area, running feature anchor points are generated at the corresponding coordinate positions, enabling multiple running feature anchor points to reflect the dynamic changing trend of engine operating condition distribution during actual vehicle operation. Furthermore, trajectory analysis is performed on multiple consecutive running feature anchor points to calculate the positional offset between adjacent running feature anchor points. ,in, This represents the offset distance of the centroid of the data distribution in the k-th window relative to the previous window; , These represent the centroid coordinates of the data distribution in the current window; , These represent the centroid coordinates of the data distribution in the previous window. When the offset directions of multiple consecutive running feature anchor points are consistent, and the offset distance exceeds a preset calibration threshold, it is determined that the current actual operating condition distribution has continuously shifted relative to the initial operating condition region, triggering the operating condition boundary calibration mechanism. At this time, the boundary offset direction is determined based on the positional relationship between the running feature anchor points and the initial multi-operating condition region boundary threshold, and the boundary adjustment amount is calculated based on the offset distance. Adaptive offset calibration is then performed on the initial multi-operating condition region boundary threshold to obtain the adaptive multi-operating condition region boundary threshold. ,in, Indicates the adaptive multi-condition region boundary threshold; This represents the initial multi-condition region boundary threshold; This represents the boundary offset calculated based on the location of the operational feature anchor point. This method enables the boundary of the operating condition region to be dynamically adjusted according to changes in the actual vehicle operating data distribution, avoiding the problem that fixed operating condition boundaries cannot adapt to different driving conditions, road environments, and vehicle load changes. This provides an updated basis for region division for the real-time operating condition identification of the subsequent engine dynamic multi-condition detector.

[0025] The engine dynamic multi-condition detector is used to perform real-time condition detection on the multi-dimensional feature vector of engine operation to determine the current time-series engine condition.

[0026] Specifically, the multi-dimensional feature vector of engine operation is input into the engine dynamic multi-condition detector. The engine dynamic multi-condition detector pre-stores condition classification rules based on adaptive multi-condition region boundary thresholds. The condition classification rules include the speed range, torque range, load characteristic range and transient change constraints corresponding to each condition region.

[0027] First, feature analysis is performed on the multi-dimensional feature vector of the engine operation at the current moment to extract engine speed features, output torque features, load features, and transient change features. Then, the current engine operating position is determined based on the current engine speed and output torque. ,in, Indicates the current engine operating position; Indicates the current engine speed; This indicates the engine output torque at the current moment. Subsequently, the current engine operating position is mapped to the speed-torque coordinate region corresponding to the universal characteristic reference map. Based on the adaptive multi-condition region boundary threshold in the engine dynamic multi-condition detector, the operating condition region to which the current operating position belongs is determined. Specifically, the current engine speed is compared with the speed range corresponding to each operating condition region, and the current engine torque is compared with the corresponding torque range. When the current engine operating position is within the speed and torque range corresponding to a certain operating condition region, that operating condition region is determined as a candidate operating condition region. Further, the load characteristic range and transient change constraint conditions corresponding to the candidate operating condition region are matched and verified with the load characteristics and transient change characteristics in the current engine operating multi-dimensional feature vector. When the current load state and change trend meet the constraint conditions corresponding to the candidate operating condition region, the candidate operating condition region is determined as the engine operating condition at the current moment. To avoid misjudgment of operating conditions due to data fluctuations at a single sampling moment, candidate operating condition results corresponding to multiple consecutive sampling moments are further obtained. The occurrence frequency of each candidate operating condition within a continuous time window is statistically analyzed, and temporal consistency is judged based on the operating condition duration and transition constraints. When the occurrence frequency of a candidate operating condition within a preset time window reaches a set threshold, the candidate operating condition is determined as the current temporal engine operating condition. When an inconsistency is detected between the current operating condition and the previous operating condition, the operating condition transition state is determined based on the direction of engine position change at adjacent moments, and the current temporal engine operating condition is updated. Through the above method, the engine dynamic multi-operating condition detector performs real-time region matching and temporal verification on the engine's multi-dimensional feature vector, achieving dynamic identification of the engine's current operating condition.

[0028] An abnormal graphics enhancement mechanism is introduced to combine the engine's universal characteristic curve and the current engine operating condition to perform graphic association mapping and layered drawing enhancement, generating layered visualization graphics of engine operating conditions.

[0029] Furthermore, generating layered visualizations of engine operating conditions includes: Based on the requirements of engine visualization tasks, a visualization task graphic element library is constructed; the engine universal characteristic curve is mapped and rendered using the visualization task graphic element library to obtain engine universal characteristic graphics, which include multi-condition regions distinguished by different background colors; an abnormal graphic enhancement mechanism is introduced to enhance the engine universal characteristic graphics by layering based on the current time-series engine operating conditions, generating engine operating condition layered visualization graphics.

[0030] First, the visualization task type is determined based on the display requirements of the engine operating condition detection results, and a visualization task graphic element library is established for different display objects. This library includes coordinate axis elements, operating condition region elements, running trajectory elements, dynamic marker elements, and anomaly enhancement elements. The coordinate axis elements define the speed-torque two-dimensional display space of the engine universal characteristic curve; the operating condition region elements represent the region range corresponding to different operating states; the running trajectory elements represent the current and historical changes in the engine's operating position; the dynamic marker elements identify the current engine operating condition position; and the anomaly enhancement elements highlight the corresponding region according to the anomaly level. Then, the engine universal characteristic curve is input into the visualization task graphic element library. Based on the mapping relationship between speed coordinates, torque coordinates, and corresponding fuel consumption characteristic parameters, different operating condition regions in the universal characteristic curve are spatially divided. The operating condition region elements are then used to render different operating condition regions, generating an engine universal characteristic graphic. The region division range is determined based on the multi-operating condition region boundary thresholds output by the engine dynamic multi-operating condition detector, ensuring that the engine universal characteristic graphic includes multi-operating condition regions corresponding to different load states. Furthermore, based on the operating condition category, speed parameters, and torque parameters corresponding to the current engine operating condition, the current engine operating position is calculated and mapped to the corresponding operating condition region in the engine universal characteristic graph, generating a dynamic operating condition display layer. Simultaneously, anomaly enhancement parameters are determined based on the deviation between the engine's multi-dimensional feature vector and the normal operating baseline features. These parameters are loaded into anomaly enhancement elements, and mark enhancement, range expansion, or hierarchical highlighting processing is performed on the corresponding region of the dynamic operating condition display layer, forming an anomaly enhancement display layer. Finally, using the engine universal characteristic graph as the base display layer, the dynamic operating condition display layer and the anomaly enhancement display layer are merged according to a preset graphic overlay order to obtain a layered visualization graph of engine operating conditions containing engine performance distribution, real-time operating status, and anomaly change information, achieving an intuitive graphical display of the engine's real-time multi-operating condition detection results.

[0031] Furthermore, an abnormal graphics enhancement mechanism is introduced to enhance the engine's universal characteristic graphics by layering them based on the current engine operating condition, generating a layered visualization of the engine operating condition, including: Using the engine universal characteristic graph as a background reference layer, the current time-series engine operating condition is mapped, rendered, and superimposed onto the background reference layer based on the visualization task graph element library to obtain a two-layer engine operating condition graph; the current engine operating position of the current time-series engine operating condition is obtained, and the current engine operating position is used as the top-level dynamic marker on the two-layer engine operating condition graph to draw a layered engine operating condition graph; an abnormal graph enhancement mechanism is introduced to enhance the layered engine operating condition graph to generate a layered engine operating condition visualization graph.

[0032] First, the engine universal characteristic graph is used as a background reference layer. This background reference layer contains the universal characteristic curve, constant-height fuel consumption line, and boundary information of different operating condition regions established based on the engine speed-torque coordinates. Then, based on the operating condition category, operating condition region, and operating status parameters corresponding to the current engine operating condition, the corresponding operating condition region drawing elements and status identifier elements are retrieved from the visualization task graphic element library to map the current engine operating condition to the corresponding coordinate region of the engine universal characteristic graph. The mapping position of the current operating condition is determined based on the current engine speed and output torque. .in, Indicates the current engine operating position; Indicates the current engine speed; This represents the engine output torque at the current moment. According to the graphics rendering rules corresponding to the current operating condition category, the current operating condition region boundary, operating condition status identifier, and operating status information are superimposed onto the background reference layer to form a two-layer engine operating condition graphics containing universal characteristic distribution information and real-time operating condition information. Further, the current engine operating position corresponding to the current time-series engine operating condition is obtained and converted into graphic coordinate points. Dynamically marked graphic elements are loaded onto the two-layer engine operating condition graphics, and these elements are updated in position according to the engine's real-time operating status. Simultaneously, operating trajectory information is generated based on the changing relationship between engine operating positions at adjacent moments, and this information is superimposed onto the two-layer engine operating condition graphics to form a layered engine operating condition graphics containing the current operating position and historical change trends. Further, an abnormal graphics enhancement mechanism is introduced to detect anomalies in the operating status of the layered engine operating condition graphics. The engine's multi-dimensional operating feature vector is compared with a preset normal operating feature range to determine the abnormal deviation amount corresponding to each feature parameter, and the degree of abnormality of the current engine state is calculated based on the abnormal deviation amount. ,in, Indicates the degree of abnormality in the current engine status; Indicates the first The abnormal weight coefficient of the item's operational characteristics; Indicates the current time. Engine operating characteristic parameters; This represents the normal reference value for the corresponding characteristic parameter. Anomalies are categorized into levels based on their severity, and corresponding enhancement elements are retrieved from the visualization task's graphical element library to enhance the display of abnormal locations in the engine operating condition layered graph. This enhancement includes expanding the abnormal area, strengthening dynamic markers, and overlaying abnormal level identifiers, enabling intuitive identification of abnormal operating states within the universal characteristic graphical space. Finally, the engine universal characteristic background reference layer, current operating condition display layer, dynamic position marker layer, and anomaly enhancement layer are merged to generate a layered visualization graph of engine operating conditions containing information on engine performance distribution, real-time operating status, and abnormal changes.

[0033] Furthermore, an abnormal graphics enhancement mechanism is introduced to enhance the engine operating condition layered graphics, generating engine operating condition layered visualization graphics, including: The engine's multi-dimensional feature vector is subjected to abnormal feature identification and classification to obtain engine abnormality level information; an abnormal graphics enhancement mechanism is introduced to anchor the engine abnormality level information onto the engine operating condition layered graphics for visualization enhancement, generating engine operating condition layered visualization graphics.

[0034] First, anomaly feature identification is performed on the engine operation multidimensional feature vector. The engine operation multidimensional feature vector at the current moment is matched with a pre-established engine normal operation benchmark feature library, which includes different speed ranges, torque ranges, and feature parameter ranges corresponding to normal operation states. For each feature parameter in the engine operation multidimensional feature vector, the deviation between the current feature value and the corresponding normal reference value is calculated to obtain the abnormal feature quantity. ,in, Indicates the first Abnormal deviations corresponding to the operational characteristic parameters of the item; Indicates the current time. Engine operating characteristic parameters; Indicates the first [stage] under normal engine operating conditions Reference values ​​for each characteristic parameter. Further, based on the importance of each abnormal characteristic parameter, anomaly weight coefficients are set, and multiple abnormal characteristic quantities are fused and calculated to obtain the engine's comprehensive anomaly evaluation value: Where A represents the engine's comprehensive anomaly evaluation value; represents the weight coefficient corresponding to the i-th abnormal feature; m represents the number of feature parameters participating in the anomaly evaluation. Based on the comparison between the comprehensive anomaly evaluation value and the preset anomaly level threshold, the current engine operating state is divided into normal, minor, moderate, and severe anomaly levels to obtain engine anomaly level information. Subsequently, an anomaly graphics enhancement mechanism is introduced, calling the corresponding graphics enhancement parameters according to the engine anomaly level information, and anchoring the anomaly level information to the region corresponding to the current engine operating position in the engine operating condition layered graphics; specifically, based on the coordinates of the current engine operating position: ,in, Indicates the current engine operating position; Indicates the current engine speed; This indicates the current engine output torque. The location coordinates are mapped to the corresponding coordinate area of ​​the engine operating condition layered graphic, and anomaly level marker elements are superimposed at that location. Simultaneously, the display range, marker shape, and enhancement level of the anomaly markers are adjusted according to the anomaly level, so that different degrees of anomaly correspond to different intensities of graphic enhancement effects. When an anomaly level is detected to increase, the display range of the corresponding anomaly area is expanded, and the display priority of the anomaly marker layer is increased; when the anomaly level decreases, the display returns to the normal operating condition state. Finally, multiple layers containing the engine's universal characteristic distribution, current operating condition, dynamic operating position, and anomaly level identifiers are merged to generate a layered visualization graphic of engine operating conditions, achieving an intuitive association between the engine's real-time operating status and anomaly change information.

[0035] Furthermore, the abnormal feature identification and classification of the multi-dimensional feature vector of engine operation are performed to obtain engine abnormality level information, including: The engine's multi-dimensional feature vector is compared with the engine's normal baseline feature library to identify abnormal engine features. A set of abnormal engine features is obtained. A preset engine anomaly assessment system is used to classify and determine the abnormal engine feature set to obtain engine anomaly level information.

[0036] Specifically, the engine normal reference feature library is established based on historical normal operating data of the engine in different speed ranges, torque ranges and load conditions. The engine normal reference feature library stores the normal feature parameter range, feature change trend and correlation between feature parameters corresponding to each operating condition.

[0037] First, the current benchmark operating condition region is determined based on the speed and torque features in the engine's multi-dimensional feature vector at the current moment. Then, the normal benchmark feature vector for the corresponding operating condition is matched from the engine's normal benchmark feature library. Each feature parameter in the current engine's multi-dimensional feature vector is compared with the normal benchmark feature vector, and the degree of deviation of each feature parameter is calculated. ,in, This indicates the degree of normalization deviation corresponding to the i-th running feature parameter; This represents the i-th engine operating characteristic parameter at the current moment; This represents the reference feature value under the corresponding operating condition in the engine's normal reference feature library; and These represent the maximum and minimum values ​​of the i-th feature parameter within the normal operating range, respectively. Further, the deviation degree corresponding to each feature parameter is compared with a preset anomaly judgment threshold. When the deviation degree of a certain feature parameter exceeds the corresponding threshold, the feature parameter is marked as an anomalous feature, and the anomalous feature type, anomalous amplitude, and anomalous occurrence location are recorded, forming an engine anomalous feature set. Where E represents the set of abnormal engine features; Indicates the type of the i-th abnormal feature; This indicates the degree of abnormal deviation; m represents the number of detected abnormal features. Subsequently, based on a preset engine anomaly assessment system, the set of engine abnormal features is graded and determined. The comprehensive anomaly evaluation value is determined according to the importance, deviation amplitude, and duration of different abnormal features. The specific calculation method is as follows: Where S represents the comprehensive evaluation value of engine anomaly; This represents the influence weight corresponding to the i-th abnormal feature; Indicates the degree of deviation of the i-th abnormal feature; Indicates the duration of the anomaly; This represents the influence coefficient of the anomaly duration. The engine anomaly evaluation value is compared with a preset anomaly level threshold. Based on the comparison result, the corresponding anomaly level is determined. When the anomaly evaluation value is below the first threshold, it is determined to be in a normal state; when the anomaly evaluation value exceeds the first threshold but does not reach the second threshold, it is determined to be in a slightly abnormal state; when the anomaly evaluation value exceeds the second threshold but does not reach the third threshold, it is determined to be in a moderately abnormal state; when the anomaly evaluation value exceeds the third threshold, it is determined to be in a severely abnormal state, thus obtaining the engine anomaly level information. Through this method, the parameter changes in the multi-dimensional feature vector of engine operation are converted into quantified anomaly level results, providing a basis for subsequent enhanced display of the engine operating condition layered graphics based on anomaly level information.

[0038] When classifying the set of engine anomalies based on a preset engine anomaly assessment system, the preset anomaly level threshold is calculated from historical normal operating samples in the engine normal baseline feature library. Specifically, the anomaly evaluation values ​​corresponding to multiple normal operating samples in the engine normal baseline feature library are obtained, and the mean anomaly evaluation and the anomaly evaluation fluctuation range under normal operating conditions are calculated based on the multiple anomaly evaluation values. , ,in, This represents the mean of the anomaly evaluations corresponding to the normally operating samples; This represents the standard deviation of anomaly evaluation for normally operating samples; Indicates the first Abnormal evaluation values ​​corresponding to samples in normal operation of the group; This represents the number of normally operating samples. The preset anomaly level threshold is determined based on the mean and standard deviation of the anomaly evaluation. The preset anomaly level threshold includes a first anomaly level threshold, a second anomaly level threshold, and a third anomaly level threshold, used to classify minor, moderate, and severe anomalies, respectively. The specific calculation is as follows: , , ,in, This indicates the threshold for the first level of abnormality. This indicates the threshold for the second level of abnormality. This represents the threshold for the third level of anomaly. Further, it represents the current engine anomaly evaluation value. Compared with the preset anomaly level threshold, when At that time, determine that the engine is at a normal operating level; when At that time, the engine was determined to be in a slightly abnormal state; when At that time, the engine was determined to be at a moderate abnormality level; when At that time, the engine is determined to be in a severe abnormality level, thereby obtaining engine abnormality level information.

[0039] In summary, the embodiments of this application have at least the following technical effects: First, a multi-dimensional real-time operating data stream of the target engine is collected, and feature extraction and calibration are performed on the multi-dimensional real-time operating data stream to generate a multi-dimensional feature vector of engine operation. Next, by combining the engine's universal characteristic curve and the actual operating characteristic data of the vehicle, adaptive boundary delineation of operating conditions is performed to construct a dynamic multi-condition engine detector. Then, the dynamic multi-condition engine detector is used to perform real-time operating condition detection on the multi-dimensional feature vector of engine operation to determine the current time-series engine operating condition. Finally, an anomaly graph enhancement mechanism is introduced to combine the engine's universal characteristic curve and the current time-series engine operating condition for graph association mapping and layered rendering enhancement, generating a layered visualization graph of engine operating conditions. This solves the technical problems of existing technologies where engine operating condition detection is difficult to adapt to real-time dynamic changes and the visualization of detection results is low, achieving the technical effect of improving the accuracy of multi-condition engine detection and the ability to intuitively display the operating status.

[0040] Example 2, based on the same inventive concept as the intuitive graphical drawing method for real-time multi-condition intelligent detection of engines in the foregoing examples, such as... Figure 3 As shown, this application provides an intuitive graphical rendering system for real-time multi-condition intelligent detection of engines, wherein the system includes: Data calibration module 11: Collects multi-dimensional real-time operating data streams of the target engine, performs feature extraction and calibration on the multi-dimensional real-time operating data streams, and generates multi-dimensional feature vectors of engine operation; Boundary division module 12: Combines the engine universal characteristic curve and the actual operating characteristic data of the vehicle to perform adaptive division of operating condition boundaries and constructs an engine dynamic multi-condition detector; Operating condition detection module 13: Uses the engine dynamic multi-condition detector to perform real-time operating condition detection on the multi-dimensional feature vectors of engine operation and determines the current time-series engine operating condition; Visualization module 14: Introduces an abnormal graphics enhancement mechanism to combine the engine universal characteristic curve and the current time-series engine operating condition to perform graphic association mapping and layered drawing enhancement, and generates layered visualization graphics of engine operating conditions.

[0041] Furthermore, the data calibration module 11 is used to perform the following method: Based on the data source types of the multidimensional real-time operating data stream, a preprocessing step is performed to obtain the multidimensional data preprocessing step process. The data source types include steady-state operating data, transient change data, and vehicle linkage data. Hierarchical preprocessing is performed on the multidimensional real-time operating data stream according to the multidimensional data preprocessing step process to obtain a standard multidimensional operating data stream. An engine multidimensional correlation feature set is extracted from the standard multidimensional operating data stream. Normalization and time-series calibration are performed on the engine multidimensional correlation feature set to generate an engine operating multidimensional feature vector.

[0042] Furthermore, the boundary delineation module 12 is used to perform the following method: Import the engine universal characteristic curve, align the coordinate system of the engine universal characteristic curve with speed as the horizontal axis and torque as the vertical axis to obtain the usable engine universal characteristic curve; anchor reference feature points, layered contour lines of fuel consumption, and full-load operating condition boundary contours on the usable engine universal characteristic curve, wherein the reference feature points include the idle speed operating point, the maximum torque operating point, and the rated power operating point; based on the reference feature points, layered contour lines of fuel consumption, and full-load operating condition boundary contours, generate a universal characteristic reference base map; project the actual vehicle operating characteristic data onto the universal characteristic reference base map to perform adaptive division of operating condition boundaries, and construct an engine dynamic multi-operating condition detector.

[0043] Furthermore, the boundary delineation module 12 is used to perform the following method: The actual vehicle operating characteristic data is projected onto the universal characteristic baseline map to generate a real vehicle operating density heatmap; the operating condition boundaries are pre-divided by combining the actual vehicle operating characteristic data and the real vehicle operating density heatmap to obtain initial multi-operating condition region boundary thresholds; a sliding calibration window is activated to adaptively calibrate the initial multi-operating condition region boundary thresholds to determine adaptive multi-operating condition region boundary thresholds; based on the adaptive multi-operating condition region boundary thresholds, the actual vehicle operating characteristic data is labeled and classified for training to construct an engine dynamic multi-operating condition detector.

[0044] Furthermore, the boundary delineation module 12 is used to perform the following method: Initiate a sliding calibration window, collect multiple real-vehicle operating data within multiple consecutive sliding calibration windows in real time, and calculate the average speed and average torque coordinates of the multiple real-vehicle operating data; determine the centroid of the data distribution of multiple windows based on the average speed and average torque coordinates; project the centroid of the data distribution of multiple windows onto the universal characteristic reference base map, and mark multiple operating feature anchor points; when a preset calibration condition is triggered, perform adaptive offset calibration on the initial multi-condition region boundary threshold based on the multiple operating feature anchor points, and determine the adaptive multi-condition region boundary threshold.

[0045] Furthermore, the visualization module 14 is used to perform the following methods: Based on the requirements of engine visualization tasks, a visualization task graphic element library is constructed; the engine universal characteristic curve is mapped and rendered using the visualization task graphic element library to obtain engine universal characteristic graphics, which include multi-condition regions distinguished by different background colors; an abnormal graphic enhancement mechanism is introduced to enhance the engine universal characteristic graphics by layering based on the current time-series engine operating conditions, generating engine operating condition layered visualization graphics.

[0046] Furthermore, the visualization module 14 is used to perform the following methods: Using the engine universal characteristic graph as a background reference layer, the current time-series engine operating condition is mapped, rendered, and superimposed onto the background reference layer based on the visualization task graph element library to obtain a two-layer engine operating condition graph; the current engine operating position of the current time-series engine operating condition is obtained, and the current engine operating position is used as the top-level dynamic marker on the two-layer engine operating condition graph to draw a layered engine operating condition graph; an abnormal graph enhancement mechanism is introduced to enhance the layered engine operating condition graph to generate a layered engine operating condition visualization graph.

[0047] Furthermore, the visualization module 14 is used to perform the following methods: The engine's multi-dimensional feature vector is subjected to abnormal feature identification and classification to obtain engine abnormality level information; an abnormal graphics enhancement mechanism is introduced to anchor the engine abnormality level information onto the engine operating condition layered graphics for visualization enhancement, generating engine operating condition layered visualization graphics.

[0048] Furthermore, the visualization module 14 is used to perform the following methods: The engine's multi-dimensional feature vector is compared with the engine's normal baseline feature library to identify abnormal engine features. A set of abnormal engine features is obtained. A preset engine anomaly assessment system is used to classify and determine the abnormal engine feature set to obtain engine anomaly level information.

[0049] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for generating intuitive graphical representations of real-time multi-condition intelligent detection of an engine, characterized in that: The method includes: Collect multidimensional real-time operation data streams of the target engine, perform feature extraction and calibration on the multidimensional real-time operation data streams, and generate multidimensional feature vectors of engine operation; By combining the universal characteristic curve of the engine and the actual operating characteristic data of the vehicle, the working condition boundary is adaptively divided, and a dynamic multi-working condition detector of the engine is constructed. The engine dynamic multi-condition detector is used to perform real-time condition detection on the multi-dimensional feature vector of engine operation to determine the current time-series engine condition. An abnormal graphics enhancement mechanism is introduced to combine the engine's universal characteristic curve and the current engine operating condition to perform graphic association mapping and layered drawing enhancement, generating layered visualization graphics of engine operating conditions.

2. The method for generating intuitive graphical representations of real-time multi-condition intelligent detection of an engine as described in claim 1, characterized in that, Feature extraction and calibration are performed on the multidimensional real-time operation data stream to generate a multidimensional feature vector of engine operation, including: Based on the data source types of the multidimensional real-time running data stream, the preprocessing steps are analyzed to obtain the multidimensional data preprocessing process. The data source types include steady-state running data, transient change data, and vehicle linkage data. The multidimensional real-time running data stream is subjected to hierarchical preprocessing according to the multidimensional data preprocessing steps to obtain a standard multidimensional running data stream. Extract the engine multidimensional correlation feature set from the standard multidimensional operating data stream; Normalization and time-series calibration are performed on the engine's multidimensional associated feature set to generate a multidimensional feature vector of engine operation.

3. The method for generating intuitive graphical representations of real-time multi-condition intelligent detection of an engine as described in claim 1, characterized in that, By combining engine universal characteristic curves and actual vehicle operating characteristic data, adaptive boundary delineation of operating conditions is performed to construct a dynamic multi-condition engine detector, including: Import the engine universal characteristic curve, align the engine universal characteristic curve with speed as the horizontal axis and torque as the vertical axis to obtain the usable engine universal characteristic curve; Anchor reference feature points, layered high fuel consumption lines, and full load condition boundary contours on the available engine universal characteristic curves, wherein the reference feature points include the idle speed condition point, the maximum torque condition point, and the rated power condition point. Based on the aforementioned benchmark feature points, layered high fuel consumption lines, and full-load working condition boundary contour lines, a universal characteristic benchmark base map is generated by overlaying. The actual operating characteristic data of the vehicle is projected onto the universal characteristic reference base map to perform adaptive division of operating condition boundaries, and an engine dynamic multi-condition detector is constructed.

4. The method for drawing intuitive graphics for real-time multi-condition intelligent detection of an engine as described in claim 3, characterized in that, The actual operating characteristic data of the vehicle is projected onto the universal characteristic baseline map for adaptive boundary division of operating conditions, and an engine dynamic multi-condition detector is constructed, including: The actual vehicle operating characteristic data is projected onto the universal characteristic reference base map to generate a real vehicle operating density heat map; By combining the actual vehicle operation characteristic data with the actual vehicle operation density heat map, the working condition boundary is pre-divided to obtain the initial multi-working condition region boundary threshold. Start the sliding calibration window to adaptively calibrate the initial multi-condition region boundary threshold, and determine the adaptive multi-condition region boundary threshold. Based on the adaptive multi-condition region boundary threshold, the actual vehicle operating feature data is labeled and classified for training to construct an engine dynamic multi-condition detector.

5. The method for generating intuitive graphical representations of real-time multi-condition intelligent detection of an engine as described in claim 4, characterized in that, Initiate a sliding calibration window to adaptively calibrate the initial multi-condition region boundary thresholds, and determine the adaptive multi-condition region boundary thresholds, including: Start the sliding calibration window, collect multiple real vehicle operation data in multiple consecutive sliding calibration windows in real time, and calculate the average speed and average torque coordinates of the multiple real vehicle operation data; Based on the average rotational speed and average torque coordinates, determine the centroid of the data distribution for multiple windows; Project the centroids of the multiple window data distributions onto the universal characteristic baseline map and mark multiple running feature anchor points; When the preset calibration condition is triggered, the initial multi-condition region boundary threshold is adaptively offset and calibrated based on the multiple operating feature anchor points to determine the adaptive multi-condition region boundary threshold.

6. The method for generating intuitive graphical representations of real-time multi-condition intelligent detection of an engine as described in claim 1, characterized in that, Generate layered visualizations of engine operating conditions, including: Based on the requirements of engine visualization tasks, a library of graphical elements for visualization tasks will be constructed. The engine universal characteristic curve is mapped and rendered using the visualization task graphic element library to obtain the engine universal characteristic graphic, which includes multi-condition areas distinguished by different background colors. An abnormal graphics enhancement mechanism is introduced to enhance the engine universal characteristic graphics by layering them based on the current engine operating conditions, generating layered visualization graphics of engine operating conditions.

7. The method for drawing intuitive graphics for real-time multi-condition intelligent detection of an engine as described in claim 6, characterized in that, An abnormal graphics enhancement mechanism is introduced to enhance the engine's universal characteristic graphics by layering them based on the current engine operating condition, generating a layered visualization of the engine operating condition, including: Using the engine universal characteristic graphic as the background reference layer, the current time-series engine operating condition is mapped, rendered, and superimposed onto the background reference layer based on the visualization task graphic element library to obtain a two-layer engine operating condition graphic. Obtain the current engine operating position of the current time-series engine operating condition, and use the current engine operating position as the top-level dynamic mark on the engine operating condition double-layer graph to draw the engine operating condition layered graph; An abnormal graphics enhancement mechanism is introduced to enhance the engine operating condition layered graphics, generating engine operating condition layered visualization graphics.

8. The method for drawing intuitive graphics for real-time multi-condition intelligent detection of an engine as described in claim 7, characterized in that, An abnormal graphics enhancement mechanism is introduced to enhance the engine operating condition layered graphics, generating engine operating condition layered visualization graphics, including: Anomaly feature identification and classification are performed on the multi-dimensional feature vector of engine operation to obtain engine anomaly level information; An abnormal graphics enhancement mechanism is introduced to anchor the engine abnormality level information onto the engine operating condition layered graphics for visualization enhancement, thereby generating an engine operating condition layered visualization graphics.

9. The method for drawing intuitive graphics for real-time multi-condition intelligent detection of an engine as described in claim 8, characterized in that, Anomaly feature identification and classification are performed on the multi-dimensional feature vector of engine operation to obtain engine anomaly level information, including: The engine's multi-dimensional feature vector is compared and identified with the engine's normal benchmark feature library under operating conditions to obtain a set of abnormal engine features. A pre-defined engine anomaly assessment system is used to classify and determine the set of engine anomaly features to obtain engine anomaly level information.

10. A real-time multi-condition intelligent detection and intuitive graphic drawing system for engines, characterized in that, The system is used to implement the intuitive graphical drawing method for real-time multi-condition intelligent detection of an engine as described in any one of claims 1-9, the system comprising: Data calibration module: Collects multi-dimensional real-time operating data streams of the target engine, performs feature extraction and calibration on the multi-dimensional real-time operating data streams, and generates multi-dimensional feature vectors of engine operation; Boundary segmentation module: Combines the engine universal characteristic curve and the actual operating characteristic data of the vehicle to perform adaptive segmentation of the working condition boundary and constructs a dynamic multi-working condition detector for the engine; Operating condition detection module: The engine dynamic multi-condition detector is used to perform real-time operating condition detection on the multi-dimensional feature vector of engine operation to determine the current time-series engine operating condition; Visualization module: Introduces an abnormal graphics enhancement mechanism to combine the engine universal characteristic curve and the current time-series engine operating condition to perform graphic association mapping and layered drawing enhancement, generating layered visualization graphics of engine operating conditions.

Images