Multi-mode parameter optimization control system and method for vehicle engine performance monitoring

Abstract

The invention discloses a multi-mode parameter optimization control system and method for vehicle engine performance monitoring, and relates to the technical field of vehicle parameter control. Vehicle engine operation parameters are monitored in real time, and a time sequence operation state set is constructed; dividing an observation period, and obtaining various types of operation parameter time sequence matrixes of the vehicle engine in the observation period; the motion state of the vehicle is determined, and the periodic offset degree of each type of operation parameters is evaluated; according to the evaluation data, constructing an inverted cone model to carry out offset hierarchy division on each type of operation parameters; respectively carrying out iterative optimization analysis on the operation parameters under each level, and outputting an optimization correction coefficient; the various types of operation parameters are corrected, and the performance regulation and control index of the corrected vehicle engine is analyzed; determining the health state of the vehicle engine by judging the performance regulation index distribution state of the vehicle engine in the observation period; according to the invention, the performance of the vehicle engine can be accurately regulated and controlled.

Description

Technical Field

[0001] This invention relates to the field of vehicle parameter control technology, specifically to a multi-mode parameter optimization control system and method for monitoring vehicle engine performance. Background Technology

[0002] As the automotive industry transforms towards energy conservation, intelligence, and low carbon emissions, the engine, as the core powertrain, has its operating efficiency, emission control, and reliability becoming the core research directions for the industry. With the increasing prevalence of vehicles in various regions, their impact on users is gradually increasing in normal social production activities. As the core power source for vehicle propulsion, the importance of engine performance testing during vehicle use is gradually increasing. Traditional engine monitoring and control modes are mostly based on single sensor parameters for safety threshold alarms, such as temperature or speed. However, this approach suffers from high monitoring error rates and parameter adjustment lags, failing to meet the needs of accurate monitoring and multi-parameter correlation analysis and optimization of the engine under various vehicle movement conditions, thus hindering engine performance optimization and control. Summary of the Invention

[0003] The purpose of this invention is to provide a multi-mode parameter optimization control system and method for monitoring vehicle engine performance, so as to solve the problems raised in the prior art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: A multi-mode parameter optimization control method for vehicle engine performance monitoring, comprising the following steps: By deploying a sensor platform, the operating parameters of the vehicle engine are monitored in real time, and a time-series operating status set is constructed. The time-series operating state set of the vehicle engine is continuously retrieved and divided into observation periods to obtain the time-series matrix of various types of operating parameters of the vehicle engine within the observation period. Determine the vehicle's motion state within the observation period, and evaluate the degree of period deviation of each type of operating parameter under the current vehicle motion state; Based on the periodic offset evaluation data of various types of operating parameters, an inverted cone model is constructed to divide the offset levels of each type of operating parameter; and based on the offset level division data of the inverted cone model, iterative optimization analysis is performed on the operating parameters under each level, and optimization correction coefficients are output. Each type of operating parameter was corrected, and the performance regulation index of the vehicle engine after correction was analyzed. The health status of the vehicle engine was determined by judging the distribution of the performance regulation index of the vehicle engine within the observation period.

[0005] Furthermore, a sensor platform is constructed by associating parameter sensors deployed on the vehicle engine; the operating parameters of the vehicle engine are monitored and collected in real time based on the sensor platform to construct a time-series operating state set; wherein, the time-series operating state set is based on a timeline as the development axis, recording the actual values ​​of various types of operating parameters of the vehicle engine at each time point. Furthermore, by dividing the observation period, the time sequence operation state set is retrieved in reverse timeline starting from the current time point to obtain the time sequence operation state set corresponding to the observation period. The time-series operation status set within the observation period is classified according to the engine operation parameter type and a mapping matrix transformation is performed to obtain the corresponding time-series matrix of each type of operation parameter; wherein, the operation parameter time-series matrix records the actual value of the corresponding type of operation parameter at each time point within the observation period.

[0006] Furthermore, the motion state of the vehicle within the observation period is determined by analyzing the speed changes of the vehicle at consecutive time points within the observation period; the test reference values ​​op(i,c) of each type of operating parameter corresponding to the vehicle's motion state are retrieved, and combined with the time series matrix of the corresponding type of operating parameter, the period offset degree Dpd(i) of each type of operating parameter is analyzed; the calculation is as follows: ; Where Dpd(i) represents the period offset of the operating parameter of type i; n(T) represents the number of time points within the observation period; op(i,t) represents the actual value of the operating parameter of type i at time t within the observation period; op(i,c) represents the test reference value of the operating parameter of type i under the current vehicle motion state within the observation period; op(i,max) and op(i,min) represent the actual maximum and minimum values ​​of the operating parameter of type i within the observation period, respectively; where i represents the type number of the engine operating parameter. It should be noted that the vehicle's motion states include the vehicle being stationary when the engine starts, the vehicle accelerating, the vehicle moving at a constant speed, and the vehicle decelerating; while the test reference values ​​for the corresponding type of operating parameters refer to the standard reference values ​​for the corresponding type of engine operating parameters at different speeds under different motion states during vehicle production testing. Furthermore, by comprehensively considering the periodic offset of various types of operating parameters, an inverted cone model is constructed to fill the model space for each type of operating parameter, thereby determining the hierarchical distribution of the corresponding type of operating parameter; the specific steps are as follows: S1. Based on the periodic offset data of each type of operating parameter, the data is sorted by descending order comparison to generate a descending axis. Parameter identity labels [i,Dpd(i)] are generated for the periodic offset data of each type of operating parameter after sorting. S2. Generate an inverted cone space by constructing an inverted cone model, and fill the bottom layer of the inverted cone space with the parameter identity label [i,Dpd(i)] corresponding to the maximum value of the period offset according to the identity labels of each parameter on the descending axis; S3. Determine the offset classification window ΔDpd, and perform window truncation on the periodic offset data of the remaining parameter identity labels on the descending axis. Parameter identity labels that meet the constraints are classified into the same level; wherein the constraints are: ; Where Dpd(i) and Dpd(i+1) are the periodic offset degrees of adjacent parameter identity labels on the descending axis; Dpd(i,q) and Dpd(i,p) are the periodic offset degrees of the parameter identity labels corresponding to the start and end points within the offset classification window, respectively. S4. On the descending axis, according to the cutting order of the offset classification window, fill the parameter identity labels of each window into the free level of the inverted cone space in turn until the parameter identity labels on the descending axis are completely cut and filled, and determine the offset level division data of the inverted cone space. Based on the offset level division data in the inverted cone model, the types and periodic offset degree data of the operating parameters filled in each offset level are considered. An optimization iterative analysis group for operating parameters is established, and the periodic offset degree data of each type of operating parameter in each offset level is imported into the optimization iterative analysis group to determine the optimization correction coefficient Otc(i,h) of each type of operating parameter at the current time. The analysis is as follows: ; Where Otc(i,h) is the optimization correction coefficient of the operating parameter of type i at time h; w(i) is the offset correction analysis weight of the operating parameter of type i; w(k) is the offset correction analysis weight of the operating parameter of type k included in the offset level; w(g) is the offset correction analysis weight of the global operating parameter of type; Dpd(i,k) is the periodic offset degree of the operating parameter of type i in the offset level of type k; i(k) is the number of operating parameters of type k included in the offset level; iA is the number of global operating parameters of type; z1 and z2 are iteration constants; where the number of global operating parameters of type corresponds to the number of operating parameters of each type of engine collected. The values ​​of the offset correction analysis weights w(i), w(k), and w(g) for each type of operating parameter, the type of operating parameters contained in the offset level, and the type of operating parameters contained globally satisfy the following conditions: ; Where m is the number of offset levels in the inverted cone model; It should be noted that the offset level labels for the inverted cone model are determined from bottom to top, and are synchronized with the filling order of the operating parameters.

[0007] Furthermore, based on the optimized correction coefficient Otc(i,h) output at the current moment for each type of operating parameter, the corresponding type of operating parameter is corrected for the next moment. The calculation is as follows: op(i,h+1)=op(i,h)×(1+Otc(i,h)); Where op(i,h+1) is the correction value of the operating parameter of type i at the current time step and the next time step; op(i,h) is the actual value of the operating parameter of type i at the current time step h. By comprehensively considering the correction values ​​of various types of operating parameters at the next moment, and comparing and analyzing the proportion of changes in operating parameters before and after correction, the performance regulation index P of the vehicle engine at the next moment is determined; the analysis is as follows: ; Determine the performance regulation judgment parameter ΔP and judge the current vehicle engine performance regulation state; when P≤ΔP, judge the current vehicle engine performance regulation as inefficient regulation; when P>ΔP, judge the current vehicle engine performance regulation as effective regulation. The system continuously monitors the performance regulation status of the vehicle engine at each time point within the observation period to determine the distribution of the vehicle engine performance regulation index at each time point within the observation period. It also coordinates the number of time points where inefficient regulation and effective regulation occur within the observation period. Based on preset abnormal regulation trigger conditions, when the trigger conditions are met, it outputs a risk warning for abnormal operation of the vehicle engine.

[0008] It should be noted that the abnormal control trigger conditions are preset by humans, and the normal state is a latent state. When the performance control of the vehicle engine meets the trigger conditions, a risk warning will be issued.

[0009] Multi-mode parameter optimization control system for vehicle engine performance monitoring: The system includes an operation parameter acquisition unit, a time series data processing unit, an operation parameter offset analysis unit, an inverted cone model construction unit, an operation parameter correction analysis unit, and an abnormal risk warning unit. The operating parameter acquisition unit constructs a sensor platform through several parameter sensors to monitor and acquire the operating parameters of the vehicle engine in real time and construct a time-series operating state set. The time-series data processing unit divides the observation period, obtains the time-series operation state set corresponding to the observation period, classifies it according to the engine operation parameter type, and performs mapping matrix transformation to obtain the time-series matrix corresponding to each type of operation parameter. The operation parameter offset analysis unit determines the vehicle's motion state within the observation period, retrieves the test reference values ​​of each type of operation parameter under the corresponding vehicle motion state, and analyzes the period offset degree of each type of operation parameter in conjunction with the time series matrix of the corresponding type of operation parameter.

[0010] The inverted cone model construction unit coordinates the periodic offset of various types of operating parameters, and fills the model space with various types of operating parameters by constructing an inverted cone model to determine the hierarchical division and distribution of the corresponding types of operating parameters. The operation parameter correction analysis unit is based on the offset level division data in the inverted cone model, and coordinates the type and period offset degree data of the operation parameters filled in each offset level; it establishes an optimization iterative analysis group for operation parameters, imports the period offset degree data of each type of operation parameter in each offset level into the optimization iterative analysis group, and determines the optimization correction coefficient of each type of operation parameter at the current moment. The abnormal risk warning unit corrects the corresponding type of operating parameters at the next moment based on the optimization correction coefficients output at the current moment for each type of operating parameter; it determines the performance control index of the vehicle engine at the next moment by comparing and analyzing the proportion of changes in operating parameters before and after correction; and it determines the health status of the vehicle engine by judging the distribution of the performance control index of the vehicle engine within the observation period.

[0011] Compared with the prior art, the beneficial effects of the present invention are: This invention collects multiple operating parameters of a vehicle engine within an observation period and analyzes the degree of period deviation. Based on the deviation of each operating parameter, an inverted cone model is constructed to classify the deviation levels of each operating parameter. This allows for iterative analysis after associating each operating parameter with global operating parameters to determine the correction coefficient of each operating parameter. By correcting the real-time engine operating parameters and analyzing the engine's performance control index before and after correction, the control state of the engine is determined. Based on the distribution of the vehicle's control state within the period, abnormal risk warnings for the vehicle engine are provided. This invention improves the performance imbalance phenomenon of traditional single-parameter control by mining the correlation of multiple operating parameters. Furthermore, by correcting parameters based on the pattern recognition of vehicle motion state, it ensures the adaptability of control strategies for multiple operating parameters under different driving scenarios, thereby improving the precision control of vehicle engine performance. Attached Figure Description

[0012] Figure 1 This is a flowchart illustrating the multi-mode parameter optimization control method for vehicle engine performance monitoring according to the present invention. Detailed Implementation

[0013] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] Example 1: As Figure 1 As shown, the present invention provides a technical solution: A multi-mode parameter optimization control method for vehicle engine performance monitoring, comprising the following steps: By deploying a sensor platform, the operating parameters of the vehicle engine are monitored in real time, and a time-series operating status set is constructed. The time-series operating state set of the vehicle engine is continuously retrieved and divided into observation periods to obtain the time-series matrix of various types of operating parameters of the vehicle engine within the observation period. Determine the vehicle's motion state within the observation period, and evaluate the degree of period deviation of each type of operating parameter under the current vehicle motion state; Based on the periodic offset evaluation data of various types of operating parameters, an inverted cone model is constructed to divide the offset levels of each type of operating parameter; and based on the offset level division data of the inverted cone model, iterative optimization analysis is performed on the operating parameters under each level, and optimization correction coefficients are output. Each type of operating parameter was corrected, and the performance regulation index of the vehicle engine after correction was analyzed. The health status of the vehicle engine was determined by judging the distribution of the performance regulation index of the vehicle engine within the observation period.

[0015] Furthermore, a sensor platform is constructed by associating parameter sensors deployed on the vehicle engine; the operating parameters of the vehicle engine are monitored and collected in real time based on the sensor platform to construct a time-series operating state set; wherein, the time-series operating state set is based on a timeline as the development axis, recording the actual values ​​of various types of operating parameters of the vehicle engine at each time point. It should be noted that, in this embodiment, the engine operating parameters include, but are not limited to, engine speed, effective torque, and fuel injection quantity; the corresponding parameter sensors include, but are not limited to, crankshaft position sensor, torque sensor, and fuel flow sensor. Furthermore, by dividing the observation period, the time sequence operation state set is retrieved in reverse time from the current time point as the starting point to obtain the time sequence operation state set corresponding to the observation period. It should be noted that the division of the observation period in this embodiment is affected by the engine's operating factors, and the changes in the operating condition data generated by its operation generally occur within seconds or even milliseconds; therefore, this factor needs to be considered when dividing the observation period. Since the data acquisition and analysis processing in the process of observing engine operating parameters, offset analysis, and iterative correction in this application are mainly used to regulate the real-time operating condition of the engine, it is mainly applicable to adjusting the dynamic operating data of the engine and is more inclined to intelligent regulation of operating condition data. Therefore, the engine operating data and analysis and regulation data monitored in its observation interval must conform to the change frequency of the engine's real-time operating condition data. If an excessively long period is selected, such as several minutes or even hours, the observation period will be too long. Although the amount of data collected is sufficient to achieve more accurate deviation analysis of operating parameters, the excessively large control time scale does not meet the real-time control requirements of the engine. Therefore, the duration of the observation period is divided into units on the order of seconds. For example, if the engine's operating data changes in the order of seconds, the observation interval can be selected to be several seconds long; if the engine's operating data changes in the order of milliseconds, the observation interval can be selected to be several milliseconds long. By selecting an appropriate observation interval, the frequency of change of the engine's instantaneous operating conditions can be better adapted, ensuring that a certain amount of operating parameter data is available for deviation analysis and optimization correction analysis, while also responding to the needs of real-time engine control. Therefore, in dividing the observation period, a time interval that better meets the needs of real-time engine operating data changes should be selected, which can be dynamically adjusted to achieve intelligent correction. The time-series operation status set within the observation period is classified according to the engine operation parameter type and a mapping matrix transformation is performed to obtain the corresponding time-series matrix of each type of operation parameter; wherein, the operation parameter time-series matrix records the actual value of the corresponding type of operation parameter at each time point within the observation period.

[0016] Furthermore, the motion state of the vehicle within the observation period is determined by analyzing the speed changes of the vehicle at consecutive time points within the observation period; the test reference values ​​op(i,c) of each type of operating parameter corresponding to the vehicle's motion state are retrieved, and combined with the time series matrix of the corresponding type of operating parameter, the period offset degree Dpd(i) of each type of operating parameter is analyzed; the calculation is as follows: ; Where Dpd(i) represents the period offset of the operating parameter of type i; n(T) represents the number of time points within the observation period; op(i,t) represents the actual value of the operating parameter of type i at time t within the observation period; op(i,c) represents the test reference value of the operating parameter of type i under the current vehicle motion state within the observation period; op(i,max) and op(i,min) represent the actual maximum and minimum values ​​of the operating parameter of type i within the observation period, respectively; where i represents the type number of the engine operating parameter. It should be noted that the vehicle's motion states include the vehicle being stationary when the engine starts, the vehicle accelerating, the vehicle moving at a constant speed, and the vehicle decelerating; while the test reference values ​​for the corresponding type of operating parameters refer to the standard reference values ​​for the corresponding type of engine operating parameters at different speeds under different motion states during vehicle production testing. In this embodiment, the specific test reference value of the corresponding type of operating parameter is determined according to the real-time speed of the vehicle under different states; if the vehicle's motion state is constant speed during the observation period, then the standard value of the corresponding engine operating parameter at the current real-time speed is determined when the vehicle is moving at a constant speed, such as the standard value of engine speed, standard value of fuel consumption, etc. It should also be noted that if the vehicle's state changes during the observation period, such as changing from constant speed to acceleration, the reference value for the test during the offset analysis of engine operating parameters is the vehicle motion state closest to the current time point. If the vehicle is initially moving at a constant speed and then changes to acceleration until the current time point during the observation period, the acceleration form is taken as the vehicle motion state corresponding to the offset analysis of engine operating parameters. Furthermore, by comprehensively considering the periodic offset of various types of operating parameters, an inverted cone model is constructed to fill the model space for each type of operating parameter, thereby determining the hierarchical distribution of the corresponding type of operating parameter; the specific steps are as follows: S1. Based on the periodic offset data of each type of operating parameter, the data is sorted by descending order comparison to generate a descending axis. Parameter identity labels [i,Dpd(i)] are generated for the periodic offset data of each type of operating parameter after sorting. S2. Generate an inverted cone space by constructing an inverted cone model, and fill the bottom layer of the inverted cone space with the parameter identity label [i,Dpd(i)] corresponding to the maximum value of the period offset according to the identity labels of each parameter on the descending axis; S3. Determine the offset classification window ΔDpd, and perform window truncation on the periodic offset data of the remaining parameter identity labels on the descending axis. Parameter identity labels that meet the constraints are classified into the same level; wherein the constraints are: ; Where Dpd(i) and Dpd(i+1) are the periodic offset degrees of adjacent parameter identity labels on the descending axis; Dpd(i,q) and Dpd(i,p) are the periodic offset degrees of the parameter identity labels corresponding to the start and end points within the offset classification window, respectively. S4. On the descending axis, according to the cutting order of the offset classification window, fill the parameter identity labels of each window into the free level of the inverted cone space in turn until the parameter identity labels on the descending axis are completely cut and filled, and determine the offset level division data of the inverted cone space. It should be noted that in this embodiment, the window truncation and inverted cone space filling of the parameter identity labels on the descending axis are performed in an alternating order. This corresponds to: after completing one offset classification window truncation on the descending axis, obtaining the currently truncated parameter identity label, and traversing the free levels in the inverted cone space from bottom to top. Based on the traversal results, the first obtained free level is selected, and the truncated parameter identity label is filled into it, thus completing one cycle of parameter identity label window truncation and inverted cone space filling. The traversal criterion for the free levels in the inverted cone space is that the corresponding free level does not contain any parameter identity labels. Based on the offset level division data in the inverted cone model, the types and periodic offset degree data of the operating parameters filled in each offset level are considered. An optimization iterative analysis group for operating parameters is established, and the periodic offset degree data of each type of operating parameter in each offset level is imported into the optimization iterative analysis group to determine the optimization correction coefficient Otc(i,h) of each type of operating parameter at the current time. The analysis is as follows: ; Where Otc(i,h) is the optimization correction coefficient of the operating parameter of type i at time h; w(i) is the offset correction analysis weight of the operating parameter of type i; w(k) is the offset correction analysis weight of the operating parameter of type k included in the offset level; w(g) is the offset correction analysis weight of the global operating parameter of type; Dpd(i,k) is the periodic offset degree of the operating parameter of type i in the offset level of type k; i(k) is the number of operating parameters of type k included in the offset level; iA is the number of global operating parameters of type; z1 and z2 are iteration constants; where the number of global operating parameters of type corresponds to the number of operating parameters of each type of engine collected. The values ​​of the offset correction analysis weights w(i), w(k), and w(g) for each type of operating parameter, the type of operating parameters contained in the offset level, and the type of operating parameters contained globally satisfy the following conditions: ; Where m is the number of offset levels in the inverted cone model; It should be noted that the offset level labeling of the inverted cone model is determined from bottom to top, and it is synchronized with the filling order of the operating parameters. For example, if the bottom layer label is 1, then when the upper layer completes the filling of the operating parameters, the generated level label will be 2; the rest are generated in the same way.

[0017] Furthermore, based on the optimized correction coefficient Otc(i,h) output at the current moment for each type of operating parameter, the corresponding type of operating parameter is corrected for the next moment. The calculation is as follows: op(i,h+1)=op(i,h)×(1+Otc(i,h)); Where op(i,h+1) is the correction value of the operating parameter of type i at the current time step and the next time step; op(i,h) is the actual value of the operating parameter of type i at the current time step h. By comprehensively considering the correction values ​​of various types of operating parameters at the next moment, and comparing and analyzing the proportion of changes in operating parameters before and after correction, the performance regulation index P of the vehicle engine at the next moment is determined; the analysis is as follows: ; Determine the performance regulation judgment parameter ΔP and judge the current vehicle engine performance regulation state; when P≤ΔP, judge the current vehicle engine performance regulation as inefficient regulation; when P>ΔP, judge the current vehicle engine performance regulation as effective regulation. The system continuously monitors the performance regulation status of the vehicle engine at each time point within the observation period to determine the distribution of the vehicle engine performance regulation index at each time point within the observation period. It also coordinates the number of time points where inefficient regulation and effective regulation occur within the observation period. Based on preset abnormal regulation trigger conditions, when the trigger conditions are met, it outputs a risk warning for abnormal operation of the vehicle engine.

[0018] It should be noted that the abnormal control trigger conditions are preset by humans and are normally in a latent state. When the performance control of the vehicle engine meets the trigger conditions, a risk warning will be issued. In this embodiment, the abnormal control triggering condition can be, but is not limited to, inefficient control of vehicle engine performance at x consecutive time points within the observation period, or a risk to vehicle health if the ratio of the number of time points with inefficient control to the number of time points with effective control within the observation period is greater than a preset value; where x is a preset constant. Example 2: The present invention provides another technical solution: Multi-mode parameter optimization control system for vehicle engine performance monitoring: The system includes an operation parameter acquisition unit, a time series data processing unit, an operation parameter offset analysis unit, an inverted cone model construction unit, an operation parameter correction analysis unit, and an abnormal risk warning unit. The operating parameter acquisition unit constructs a sensor platform through several parameter sensors to monitor and acquire the operating parameters of the vehicle engine in real time and construct a time-series operating state set. The time-series data processing unit divides the observation period, obtains the time-series operation state set corresponding to the observation period, classifies it according to the engine operation parameter type, and performs mapping matrix transformation to obtain the time-series matrix corresponding to each type of operation parameter. The operation parameter offset analysis unit determines the vehicle's motion state within the observation period, retrieves the test reference values ​​of each type of operation parameter under the corresponding vehicle motion state, and analyzes the period offset degree of each type of operation parameter in conjunction with the time series matrix of the corresponding type of operation parameter.

[0019] The inverted cone model construction unit coordinates the periodic offset of various types of operating parameters, and fills the model space with various types of operating parameters by constructing an inverted cone model to determine the hierarchical division and distribution of the corresponding types of operating parameters. The operation parameter correction analysis unit is based on the offset level division data in the inverted cone model, and coordinates the type and period offset degree data of the operation parameters filled in each offset level; it establishes an optimization iterative analysis group for operation parameters, imports the period offset degree data of each type of operation parameter in each offset level into the optimization iterative analysis group, and determines the optimization correction coefficient of each type of operation parameter at the current moment. The abnormal risk warning unit corrects the corresponding type of operating parameters at the next moment based on the optimization correction coefficients output at the current moment for each type of operating parameter; it determines the performance control index of the vehicle engine at the next moment by comparing and analyzing the proportion of changes in operating parameters before and after correction; and it determines the health status of the vehicle engine by judging the distribution of the performance control index of the vehicle engine within the observation period.

[0020] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A multi-mode parameter optimization control method for monitoring vehicle engine performance, characterized in that: By deploying a sensor platform, the operating parameters of the vehicle engine are monitored in real time, and a time-series operating status set is constructed. The time-series operating state set of the vehicle engine is continuously retrieved and divided into observation periods to obtain the time-series matrix of various types of operating parameters of the vehicle engine within the observation period. Determine the vehicle's motion state within the observation period, and evaluate the degree of period deviation of each type of operating parameter under the current vehicle motion state; Based on the periodic offset evaluation data of various types of operating parameters, an inverted cone model is constructed to classify the offset levels of various types of operating parameters. Based on the offset hierarchy data of the inverted cone model, the operating parameters under each hierarchy are iteratively optimized and analyzed, and the optimization correction coefficients are output. Each type of operating parameter was corrected, and the performance regulation index of the vehicle engine after correction was analyzed. The health status of the vehicle engine was determined by judging the distribution of the performance regulation index of the vehicle engine within the observation period.

2. The multi-mode parameter optimization control method for vehicle engine performance monitoring according to claim 1, characterized in that: The motion state of the vehicle within the observation period is determined by analyzing the speed changes of the vehicle at consecutive time points within the observation period. The test reference values ​​op(i,c) of each type of operating parameter corresponding to the vehicle's motion state are retrieved, and combined with the time series matrix of the corresponding type of operating parameter, the period offset degree Dpd(i) of each type of operating parameter is analyzed; the calculation is as follows: ； Wherein, Dpd(i) is the period offset degree of the operating parameter of type i; n(T) is the number of time points within the observation period; op(i,t) is the actual value of the operating parameter of type i at time t within the observation period; op(i,c) is the test reference value of the operating parameter of type i under the current vehicle motion state within the observation period; op(i,max) and op(i,min) are the actual maximum and minimum values ​​of the operating parameter of type i within the observation period, respectively.

3. The multi-mode parameter optimization control method for vehicle engine performance monitoring according to claim 1, characterized in that: By taking into account the periodic offset of various types of operating parameters, and filling the model space with various types of operating parameters by constructing an inverted cone model, the hierarchical division and distribution of the corresponding types of operating parameters can be determined. The analysis steps are as follows: S1. Based on the periodic offset data of each type of operating parameter, the data is sorted by descending order comparison to generate a descending axis. Parameter identity labels [i,Dpd(i)] are generated for the periodic offset data of each type of operating parameter after sorting. S2. Generate an inverted cone space by constructing an inverted cone model, and fill the bottom layer of the inverted cone space with the parameter identity label [i,Dpd(i)] corresponding to the maximum value of the period offset according to the identity labels of each parameter on the descending axis; S3. Determine the offset classification window ΔDpd, and perform window truncation on the periodic offset data of the remaining parameter identity labels on the descending axis. Parameter identity labels that meet the constraints are classified into the same level; wherein the constraints are: ； Where Dpd(i) and Dpd(i+1) are the periodic offset degrees of adjacent parameter identity labels on the descending axis; Dpd(i,q) and Dpd(i,p) are the periodic offset degrees of the parameter identity labels corresponding to the start and end points within the offset classification window, respectively. S4. On the descending axis, according to the cutting order of the offset classification window, fill the parameter identity labels of each window into the free level of the inverted cone space in turn until the parameter identity labels on the descending axis are completely cut and filled, and determine the offset level division data of the inverted cone space.

4. The multi-mode parameter optimization control method for vehicle engine performance monitoring according to claim 1, characterized in that: Based on the offset level division data in the inverted cone model, the types and periodic offset degree data of the operating parameters filled in each offset level are considered. An optimization iterative analysis group for operating parameters is established, and the periodic offset degree data of each type of operating parameter in each offset level is imported into the optimization iterative analysis group to determine the optimization correction coefficient Otc(i,h) of each type of operating parameter at the current time. The analysis is as follows: ； Where Otc(i,h) is the optimization correction coefficient of the operating parameter of type i at time h; w(i) is the offset correction analysis weight of the operating parameter of type i; w(k) is the offset correction analysis weight of the operating parameter of type k included in the offset level; w(g) is the offset correction analysis weight of the global operating parameter of type; Dpd(i,k) is the periodic offset degree of the operating parameter of type i in the offset level of type k; i(k) is the number of operating parameters of type k included in the offset level; iA is the number of global operating parameters of type; z1 and z2 are iteration constants; Among them, the values ​​of the offset correction analysis weights w(i), w(k), and w(g) for each type of operating parameter, the type of operating parameters contained in the offset level, and the type of operating parameters contained globally satisfy the following conditions: ； Where m is the number of offset levels in the inverted cone model.

5. The multi-mode parameter optimization control method for vehicle engine performance monitoring according to claim 1, characterized in that: Based on the optimized correction coefficient Otc(i,h) output at the current moment for each type of operating parameter, the corresponding type of operating parameter is corrected at the next moment. The calculation is as follows: op(i,h+1)=op(i,h)×(1+Otc(i,h)); Where op(i,h+1) is the correction value of the operating parameter of type i at the current time step and the next time step; op(i,h) is the actual value of the operating parameter of type i at the current time step h. By comprehensively considering the correction values ​​of various types of operating parameters at the next moment, and comparing and analyzing the proportion of changes in operating parameters before and after correction, the performance regulation index P of the vehicle engine at the next moment is determined; the analysis is as follows: ； Where iA represents the number of global type runtime parameters.

6. The multi-mode parameter optimization control method for vehicle engine performance monitoring according to claim 5, characterized in that: Determine the performance regulation judgment parameter ΔP and judge the current vehicle engine performance regulation state; when P≤ΔP, judge the current vehicle engine performance regulation as inefficient regulation. When P > ΔP, the current vehicle engine performance regulation is determined to be effective. The performance regulation status of the vehicle engine at each time point within the observation period is continuously monitored to determine the distribution of the performance regulation index of the vehicle engine at each time point within the observation period; and the number of time points where inefficient regulation and effective regulation occur are also considered within the observation period. By setting up pre-defined abnormal control trigger conditions, when the trigger conditions are met, a risk warning for abnormal operation of the vehicle engine is output.

7. The multi-mode parameter optimization control method for vehicle engine performance monitoring according to claim 1, characterized in that: A sensor platform is constructed by associating parameter sensors deployed on the vehicle engine; the operating parameters of the vehicle engine are monitored and collected in real time based on the sensor platform, and a time-series operating state set is constructed.

8. The multi-mode parameter optimization control method for vehicle engine performance monitoring according to claim 1, characterized in that: By dividing the observation period, the time sequence operation state set is retrieved in reverse timeline starting from the current time point to obtain the time sequence operation state set corresponding to the observation period. The time-series operating state set within the observation period is classified according to the engine operating parameter type and a mapping matrix transformation is performed to obtain the corresponding time-series matrix of each type of operating parameter.

9. A system for executing the multi-mode parameter optimization control method for vehicle engine performance monitoring as described in any one of claims 1-8, characterized in that: The system includes an operation parameter acquisition unit, a time series data processing unit, an operation parameter offset analysis unit, an inverted cone model construction unit, an operation parameter correction analysis unit, and an abnormal risk warning unit. The operating parameter acquisition unit constructs a sensor platform through several parameter sensors to monitor and acquire the operating parameters of the vehicle engine in real time and construct a time-series operating state set. The time-series data processing unit divides the observation period, obtains the time-series operation state set corresponding to the observation period, classifies it according to the engine operation parameter type, and performs mapping matrix transformation to obtain the time-series matrix corresponding to each type of operation parameter. The operation parameter offset analysis unit determines the vehicle's motion state within the observation period, retrieves the test reference values ​​of each type of operation parameter under the corresponding vehicle motion state, and analyzes the period offset degree of each type of operation parameter in conjunction with the time series matrix of the corresponding type of operation parameter.

10. The system according to claim 9, characterized in that: The inverted cone model construction unit coordinates the periodic offset of various types of operating parameters, and fills the model space with various types of operating parameters by constructing an inverted cone model to determine the hierarchical division and distribution of the corresponding types of operating parameters. The operation parameter correction analysis unit is based on the offset level division data in the inverted cone model, and coordinates the type and period offset degree data of the operation parameters filled in each offset level; it establishes an optimization iterative analysis group for operation parameters, imports the period offset degree data of each type of operation parameter in each offset level into the optimization iterative analysis group, and determines the optimization correction coefficient of each type of operation parameter at the current moment. The abnormal risk warning unit performs correction processing on the corresponding type of operating parameters at the next moment based on the optimization correction coefficient output by each type of operating parameter at the current moment. By comparing and analyzing the percentage changes in operating parameters before and after correction, the performance control index of the vehicle engine at the next moment is determined; and by judging the distribution of the performance control index of the vehicle engine within the observation period, the health status of the vehicle engine is determined.

Images