A method and system for online monitoring of system performance of unmanned heavy-duty vehicles

Abstract

The present invention provides an online monitoring method and system for the system performance of an unmanned heavy-duty vehicle, and belongs to the technical field of heavy-duty vehicle motion planning and control. The present invention takes into account the operating environment, especially the influence of the road environment on the system performance of the unmanned heavy-duty vehicle, and measures the system response transfer function of the unmanned heavy-duty vehicle under different operating environments and different typical working conditions. At the same time, a good performance knowledge base and a performance degradation knowledge base are constructed, and the characteristic parameters of the system response transfer function and production data are used as the basis for whether the system needs maintenance intervention, thereby realizing automatic online monitoring of unmanned heavy-duty vehicles.

Description

Technical Field

[0001] The present invention belongs to the technical field of heavy-duty vehicle motion planning and control, and in particular relates to an online monitoring method and system for the system performance of an unmanned heavy-duty vehicle. Background Art

[0002] Due to the harsh working environment of heavy-duty mining vehicles, large load, poor road conditions, and long working hours, the vehicle system will inevitably experience performance degradation. This performance degradation will pose a potential safety risk to the operation of heavy-duty mining vehicles. Since the driver of a traditional manually driven vehicle is always in the control loop, the evaluation of system performance is almost real-time, and the driver can keenly perceive the degradation of vehicle performance, so as to take corresponding measures: when the performance degradation is not obvious, no maintenance is required, but the amplitude of the control command is appropriately adjusted (often increased) to enable the system to complete the corresponding function; when the performance degradation is obvious, maintenance is carried out, and the driver evaluates the maintenance effect to ensure that the system function is maintained within a relatively good range. Therefore, manual driving has obvious advantages in monitoring system degradation. Unlike manual driving, since unmanned heavy-duty mining vehicles have eliminated the driver, in order to achieve safe operation, it is necessary to monitor the system performance degradation online. However, current research focuses more on monitoring when a fault occurs and subsequent fault-tolerant control, but less on system performance degradation.

[0003] Moreover, at present, the maintenance of unmanned heavy-duty mining vehicles is often carried out in a fixed cycle, such as fixed working hours, fixed mileage, etc. However, since the timing of maintenance intervention for a single vehicle does not strictly follow the above cycle, intervention too early may lead to unnecessary time, manpower and financial expenditures, and intervention too late may cause safety risks. Therefore, this fixed-cycle maintenance method has obvious disadvantages. A more reasonable way is to carry out appropriate maintenance intervention according to the actual situation of each unmanned heavy-duty mining vehicle, which requires online monitoring of the system performance of unmanned heavy-duty mining vehicles. The existing technology lacks such an online monitoring solution.

[0004] In summary, it is necessary to provide an online monitoring method and system for the system performance of an unmanned heavy-duty vehicle to solve the above problems. Summary of the invention

[0005] The present invention provides a method and system for online monitoring of the system performance of an unmanned heavy-duty vehicle. The present invention takes into account the operating environment, especially the impact of the road environment on the system performance of the unmanned heavy-duty vehicle, measures the system response transfer function of the unmanned heavy-duty vehicle under different road environments and different typical working conditions, and simultaneously constructs a good performance knowledge base and a performance attenuation knowledge base. The characteristic parameters of the system response transfer function are used as the basis for whether the system should be repaired or not, thereby realizing automatic online monitoring of the unmanned heavy-duty vehicle, which can effectively solve at least one technical problem in the background technology.

[0006] In order to solve the above-mentioned technical problems, the present invention is achieved as follows:

[0007] A method for online monitoring of system performance of an unmanned heavy-duty vehicle comprises the following steps:

[0008] Step S1, extracting typical operating conditions of each system of an unmanned heavy-duty vehicle and the duration of each typical operating condition from historical data;

[0009] Step S2, providing an unmanned heavy-duty vehicle with good system performance, defined as a first type of vehicle; providing an unmanned heavy-duty vehicle that just needs maintenance intervention, defined as a second type of vehicle; testing the system response of each system of the first type of vehicle and the second type of vehicle under typical working conditions according to different operating environments, respectively calculating the transfer function of the response of each system under different operating environments and different typical working conditions with respect to the input of the system, extracting characteristic parameters of each system response transfer function, constructing a good performance knowledge base with the characteristic parameters of the system response transfer function of the first type of vehicle and production data; constructing a performance degradation database with the characteristic parameters of the system response transfer function of the second type of vehicle and production data;

[0010] Step S3, for online monitoring of the system performance of any unmanned heavy-duty vehicle, first obtain the operating environment in which the unmanned heavy-duty vehicle is running, then use the upper limit of the duration of each typical working condition extracted in step S1 as the length of the sliding time window, monitor the system input and system response of each system of the unmanned heavy-duty vehicle online, and use the mutual correlation coefficient method to match the system input of any system of the unmanned heavy-duty vehicle with the system input under the typical working condition. If the match is successful, save the system input and system response collected in the current sliding time window, use the least squares method to fit the transfer function of the system response with respect to the system input, and extract the characteristic parameters of the transfer function;

[0011] Step S4, using the least squares method to compare the characteristic parameters obtained in step S3 with the characteristic parameters corresponding to each system under typical working conditions matched in the good performance knowledge base. If the characteristic parameters obtained in step S3 are greater than the upper limit of the characteristic parameter value range under typical working conditions in the good performance knowledge base, the characteristic parameters are deemed to be activated; at the same time, the characteristic parameters corresponding to each system under typical working conditions matched in the performance decay database are compared. If the characteristic parameters obtained in step S3 are greater than the lower limit of the characteristic parameter value range under typical working conditions in the performance decay database, the characteristic parameters are deemed to be activated; the number of activated characteristic parameters is counted, and if the number of activated characteristic parameters exceeds a preset first threshold, an alarm signal is issued that the corresponding system needs to perform performance evaluation; in addition, the current production data is compared with the production data stored in the good performance knowledge base and the performance decay knowledge base, and the total number of times the current production data is greater than the production data in the good performance knowledge base and the performance decay knowledge base is counted. If the total number is greater than a preset second threshold, an alarm signal is also issued that the corresponding system needs to perform performance evaluation;

[0012] Step S5, perform a performance evaluation on the system targeted by the alarm signal. If the evaluation result is that the scrap condition has not been reached and maintenance intervention is required, the unmanned heavy-duty vehicle is classified as a second type of vehicle, and step S2 is executed to update the performance degradation database; then the system is repaired, and the unmanned heavy-duty vehicle after the performance is restored is classified as a first type of vehicle, and step S2 is executed to update the good performance knowledge base, and a maintenance record is added at the same time; if the evaluation result is that the scrap condition has not been reached and the time for maintenance intervention has not come, the characteristic parameter values ​​under typical working conditions in the good performance knowledge base are added. The upper and lower limits of the interval are taken to reduce the upper and lower limits of the interval of characteristic parameter values ​​under typical working conditions in the performance attenuation database; if the evaluation result is that the system is completely damaged, further evaluate whether it is necessary to replace the system. If necessary, replace the system and classify the unmanned heavy-duty vehicle after performance restoration as a first type of vehicle, execute step S2, and update the good performance knowledge base; if it is assessed that the unmanned heavy-duty vehicle has reached the scrap condition, exit the online monitoring of the unmanned heavy-duty vehicle and scrap the unmanned heavy-duty vehicle.

[0013] As a preferred improvement, the unmanned heavy-load vehicle is an unmanned heavy-load vehicle for mining, and the monitoring system includes a steering system in normal mode, a braking system, a drive system in normal mode, and a lifting system, wherein: typical operating conditions of the steering system include obstacle avoidance conditions, U-turn conditions, and right-angle turning conditions at various vehicle speeds; typical operating conditions of the braking system include different brake pedal opening conditions at various initial vehicle speeds under straight roads, downhill roads, and uphill roads, and under different load conditions; typical operating conditions of the drive system include different accelerator pedal opening conditions at various initial vehicle speeds under straight roads, downhill roads, and uphill roads, and under different load conditions; typical operating conditions of the lifting system include conditions of lifting cargo boxes of different load masses according to different lifting times.

[0014] As a preferred improvement, the first type of vehicles includes brand new unmanned heavy-duty vehicles and unmanned heavy-duty mining vehicles whose performance has been restored after repair.

[0015] As a preferred improvement, the operating environment includes weather, temperature and road environment, wherein the road environment includes dry road surface, wet road surface and frozen road surface.

[0016] As a preferred improvement, the system input of the steering system is the steering wheel angle; the system response of the steering system includes the yaw angular velocity, lateral acceleration, steering angle, steering system oil pressure and steering hydraulic lever stroke; the system input of the braking system is the brake pedal opening; the system response of the braking system includes longitudinal acceleration, longitudinal speed, brake hydraulic cylinder pressure and wheel speed; the system input of the drive system is the accelerator pedal opening; the system response of the drive system includes longitudinal acceleration, longitudinal speed, drive motor torque and wheel speed; the system input of the lifting system is the lifting time; the system response of the lifting system includes the bucket lifting angle, lifting hydraulic cylinder pressure and lifting hydraulic cylinder stroke.

[0017] As a preferred improvement, the transfer functions of the steering system and the drive system are fitted using an underdamped second-order system transfer function model. The underdamped second-order system transfer function is expressed as:

[0018] ;

[0019] Where s represents the complex frequency variable in Laplace transform; represents the natural frequency of the system, represents the system damping ratio and satisfies ;

[0020] The transfer functions of the braking system and the lifting system are fitted using a first-order inertial system transfer function model. The first-order inertial system transfer function is expressed as:

[0021] ;

[0022] In the formula, Represents the system time constant.

[0023] As a preferred improvement, the characteristic parameters of the transfer function of the steering system and the drive system include: damped oscillation frequency , Rise time , Peak Time , Overshoot , Adjust time As well as the system damping ratio, the definitions of the characteristic parameters are as follows:

[0024] ;

[0025] ;

[0026] ;

[0027] ;

[0028] ;

[0029] The characteristic parameters of the transfer function of the braking system and the lifting system include: system time constant , first-order system rise time And the first-order system adjustment time , where the first-order system rise time is It represents the time for the system to go from 10% to 90% in step response. ; First-order system adjustment time .

[0030] As a preferred improvement, the good performance knowledge base and the performance degradation knowledge base have the same data structure, both adopting a tree structure, each system corresponds to a first-level branch, each first-level branch is provided with at least one second-level branch, each second-level branch corresponds to a maintenance status, and the data before maintenance is stored in the performance degradation knowledge base; the data after maintenance is stored in the good performance knowledge base; each second-level branch is provided with at least one third-level branch, each third-level branch corresponds to a road environment; each third-level branch is provided with at least one fourth-level branch, each fourth-level branch corresponds to a typical working condition; each fourth-level branch is provided with at least one fifth-level branch, each fifth-level branch corresponds to a characteristic parameter of a system response transfer function.

[0031] As a preferred improvement, the brand new state of the unmanned heavy-duty vehicle is taken as the initial node, each maintenance state corresponds to a node, and the production data includes the total mass of different types of goods transported by the unmanned heavy-duty vehicle between two adjacent nodes, the corresponding mileage, and the accumulated work done in the vertical direction to overcome gravity; in the brand new state, the production data of the unmanned heavy-duty vehicle is 0.

[0032] An unmanned heavy-load vehicle system performance online monitoring system is used for the above-mentioned unmanned heavy-load vehicle system performance online monitoring method.

[0033] The beneficial effects of the present invention are:

[0034] (1) The present invention takes into account the impact of the operating environment on the performance of unmanned heavy-duty mining vehicles, and can effectively avoid false alarms caused by changes in the operating environment;

[0035] (2) The good performance database established by the present invention includes data on the new state of the unmanned heavy-duty mining vehicle and the performance recovery state after maintenance, thus avoiding the pollution of the knowledge base by the system performance attenuation data;

[0036] (3) The monitoring method provided by the present invention can be applied to steering systems, braking systems, drive systems and lifting systems, and has good adaptability;

[0037] (4) The good performance knowledge base and the performance degradation knowledge base established by the present invention can continue to increase with the increase of the operating time of the unmanned heavy-duty mining vehicle, and can provide rich data support for the design and development of unmanned heavy-duty mining vehicles. DETAILED DESCRIPTION

[0038] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] The present invention provides an online monitoring method for the system performance of an unmanned heavy-duty vehicle, comprising the following steps:

[0040] Step S1, extracting typical operating conditions of each system of an unmanned heavy-duty vehicle and the duration of each typical operating condition from historical data.

[0041] In this embodiment, the unmanned heavy-load vehicle is an unmanned heavy-load vehicle for mining, and its system includes a steering system, a braking system, a drive system and a lifting system. The monitored system includes a steering system in normal mode, a braking system, a drive system in normal mode and a lifting system.

[0042] For the steering system, there are multiple steering modes, such as multi-axis steering system and distributed independent steering system. This solution only focuses on the response of the system in normal mode under various working conditions, without considering the activation of the system in fault-tolerant mode. This is because for unmanned heavy-duty vehicles with multiple steering modes, the fault-tolerant mode can ideally make the system state respond well, but the actual system has already experienced performance degradation or partial failure. Therefore, once the unmanned heavy-duty vehicle system performance online monitoring system detects that the fault-tolerant mode is activated, maintenance intervention should be immediately carried out.

[0043] For the drive system, there are multiple drive modes, such as distributed independent drive systems. This solution only focuses on the response of the system in normal mode under various working conditions, without considering the activation of the system in fault-tolerant mode. This is because for unmanned heavy-duty vehicles with multiple drive modes, the fault-tolerant mode can ideally also make the system state respond well, but the actual system has already experienced performance degradation or partial failure. Therefore, once the unmanned heavy-duty vehicle system performance online monitoring system detects that the fault-tolerant mode is activated, maintenance intervention should be immediately carried out.

[0044] The typical operating conditions of the steering system include obstacle avoidance conditions, U-turn conditions and right-angle turning conditions at various vehicle speeds; the typical operating conditions of the braking system include different brake pedal opening conditions at various initial vehicle speeds under straight roads, downhill roads and uphill roads, and under different load conditions; the typical operating conditions of the drive system include different accelerator pedal opening conditions at various initial vehicle speeds under straight roads, downhill roads and uphill roads, and under different load conditions; the typical operating conditions of the lifting system include the conditions of lifting cargo boxes with different load masses according to different lifting times.

[0045] Step S2, providing an unmanned heavy-duty vehicle with good system performance, defined as a first type of vehicle; providing an unmanned heavy-duty vehicle that just needs maintenance intervention, defined as a second type of vehicle; testing the system response of each system of the first type of vehicle and the second type of vehicle under typical operating conditions according to different operating environments, respectively calculating the transfer function of the system response of each system under different operating environments and different typical operating conditions with respect to the system input, extracting the characteristic parameters of the system response transfer function, and constructing a good performance knowledge base with the characteristic parameters of the system response transfer function of the first type of vehicle and production data; and constructing a performance attenuation database with the characteristic parameters of the system response transfer function of the second type of vehicle and production data.

[0046] The first type of vehicles includes brand new unmanned heavy-duty vehicles and unmanned heavy-duty vehicles whose performance has been restored after repair.

[0047] The operating environment includes weather, temperature and road environment, wherein the road environment is divided into dry road, wet road and frozen road.

[0048] The system input of the steering system is the steering wheel angle; the system response of the steering system includes yaw rate, lateral acceleration, steering angle, steering system oil pressure and steering hydraulic lever stroke.

[0049] The system input of the braking system is the brake pedal opening; the system response of the braking system includes longitudinal acceleration, longitudinal speed, brake hydraulic cylinder pressure and wheel speed.

[0050] The system input of the driving system is the accelerator pedal opening; the system response of the driving system includes longitudinal acceleration, longitudinal speed, driving motor torque and wheel speed.

[0051] The system input of the lifting system is the lifting time; the system response of the lifting system includes the lifting angle of the truck bucket, the pressure of the lifting hydraulic cylinder and the stroke of the lifting hydraulic cylinder.

[0052] For any operating environment, multiple tests are required, and the final system response is determined by taking the average value after eliminating discrete points to ensure the accuracy of the test results and avoid accidental errors.

[0053] The transfer functions of the steering system and the drive system are fitted using an underdamped second-order system transfer function model. The underdamped second-order system transfer function is expressed as:

[0054] ;

[0055] Where s represents the complex frequency variable in Laplace transform; represents the natural frequency of the system, represents the system damping ratio and satisfies ;

[0056] The transfer functions of the braking system and the lifting system are fitted using a first-order inertial system transfer function model. The first-order inertial system transfer function is expressed as:

[0057] ;

[0058] In the formula, Represents the system time constant.

[0059] The process of fitting the underdamped second-order system transfer function model and the first-order inertial system transfer function model belongs to the prior art in the field, and will not be described in detail in this embodiment.

[0060] The characteristic parameters of the transfer function of the steering system and the drive system include: damped oscillation frequency , Rise time , Peak Time , Overshoot , Adjust time As well as the system damping ratio, the definitions of the characteristic parameters are as follows:

[0061] ;

[0062] ;

[0063] ;

[0064] ;

[0065] ;

[0066] The characteristic parameters of the transfer function of the braking system and the lifting system include: system time constant , first-order system rise time And the first-order system adjustment time , where the first-order system rise time is It represents the time for the system to go from 10% to 90% in step response. ; First-order system adjustment time .

[0067] The good performance knowledge base and the performance degradation knowledge base have the same data structure, both adopting a tree structure, each system corresponds to a first-level branch, each first-level branch is provided with at least one second-level branch, each second-level branch corresponds to a maintenance state, and the data before maintenance is stored in the performance degradation knowledge base; the data after maintenance is stored in the good performance knowledge base; each second-level branch is provided with at least one third-level branch, each third-level branch corresponds to a road surface environment; each third-level branch is provided with at least one fourth-level branch, each fourth-level branch corresponds to a typical working condition; each fourth-level branch is provided with at least one fifth-level branch, each fifth-level branch corresponds to a characteristic parameter of a system response transfer function.

[0068] The new state of the unmanned heavy-duty vehicle is taken as the initial node, and each maintenance state corresponds to a node. The production data includes the total mass of different types of goods transported by the unmanned heavy-duty vehicle between two adjacent nodes, the corresponding mileage, and the accumulated work done in the vertical direction to overcome gravity. In the new state, the production data of the unmanned heavy-duty vehicle is 0.

[0069] It can be understood that the good performance knowledge base and the performance decay database are continuously updated as the life cycle of the unmanned heavy-duty vehicle progresses. In the brand new state (initial node), the characteristic parameters and production data (0) of the unmanned heavy-duty vehicle are stored in the good performance knowledge base to reflect the performance of the unmanned heavy-duty vehicle in the brand new state; as the unmanned heavy-duty vehicle is used, the production data gradually accumulates, and at the same time, each system has a certain degree of wear and tear. When the degree of wear reaches the degree of maintenance, it comes to the first maintenance node, and the production data generated from the initial node to the first maintenance node is stored in the performance decay database. At the same time, the unmanned heavy-duty vehicle is tested to obtain the characteristic parameters before the first maintenance, which are stored in the performance decay knowledge base to characterize the performance decay of the unmanned heavy-duty vehicle before the first maintenance; after the maintenance is completed, the performance of each system of the unmanned heavy-duty vehicle is restored, but there may still be a certain gap with the brand new state, so the unmanned heavy-duty vehicle after the first maintenance is retested to obtain the characteristic parameters and production data of the performance recovery after the first maintenance, which are stored in the good performance knowledge base. The above process is repeated, the data before each maintenance is stored in the performance degradation knowledge base, and the data after each maintenance is stored in the good performance knowledge base, until the unmanned heavy-duty vehicle is scrapped, and the performance status data of the unmanned heavy-duty vehicle throughout its life cycle is obtained.

[0070] Step S3, for online monitoring of the system performance of any unmanned heavy-duty vehicle, first obtain the operating environment in which the unmanned heavy-duty vehicle is operating, then use the upper limit of the duration of each typical operating condition extracted in step S1 as the length of the sliding time window, monitor the system input and system response of each system of the unmanned heavy-duty vehicle online, and use the mutual correlation coefficient method to match the system input of any system of the unmanned heavy-duty vehicle with the system input under typical operating conditions. If the match is successful, save the system input and system response collected in the current sliding time window, use the least squares method to fit the transfer function of the system response with respect to the system input and extract the characteristic parameters of the transfer function.

[0071] The temperature can be obtained by querying the weather forecast website and fusing the measurements of the vehicle-mounted sensors. Preferably, the temperature sensor is used to directly obtain the measurement.

[0072] The acquisition of weather and road conditions can be achieved through vehicle-mounted sensors, preferably visual sensors. The data collected by the visual sensors are sent to a convolutional neural network for processing, and the prediction results of the weather and road conditions are output.

[0073] Whether the system input of any system of the unmanned heavy-duty vehicle matches the system input under the typical working conditions of the current operating environment can be judged by the mutual correlation coefficient. When the mutual correlation coefficient is greater than a preset threshold, the match is considered successful. In this embodiment, the preset threshold value is 0.7. In other embodiments, the value of the preset threshold value can also be adjusted according to actual conditions.

[0074] Mutual Correlation Coefficient R Calculated by the following formula:

[0075] ;

[0076] In the formula, Indicates the time window i Sampling time ， System input for any system; represents its mean; Represents the system input under typical conditions of the current operating environment, represents its mean; n Indicates the total number of samples in the time window.

[0077] Step S4, using the least squares method to compare the characteristic parameters obtained in step S3 with the characteristic parameters corresponding to each system under typical working conditions matched in the good performance knowledge base. If the characteristic parameters obtained in step S3 are greater than the upper limit of the characteristic parameter value range under typical working conditions in the good performance knowledge base, the characteristic parameters are deemed to be activated; at the same time, the characteristic parameters corresponding to each system under typical working conditions matched in the performance decay database are compared. If the characteristic parameters obtained in step S3 are greater than the lower limit of the characteristic parameter value range under typical working conditions in the performance decay database, the characteristic parameters are deemed to be activated; the number of activated characteristic parameters is counted, and if the number of activated characteristic parameters exceeds a preset first threshold, an alarm signal is issued that the corresponding system needs to perform performance evaluation; in addition, the current production data is compared with the production data stored in the good performance knowledge base and the performance decay knowledge base, and the total number of times the current production data is greater than the production data in the good performance knowledge base and the performance decay knowledge base is counted. If the total number is greater than a preset second threshold, an alarm signal is also issued that the corresponding system needs to perform performance evaluation.

[0078] For example, for a lifting system with good performance, the time interval to reach the target lifting height is [26s, 30s]; if the current lifting system reaches the same target lifting height in =29s, which is less than the upper limit of the time interval, then the characteristic parameter is not activated in the good performance knowledge base; for the lifting system with performance degradation, since the lifting system has reached the point where it needs maintenance after working for a long time, the time interval to reach the same target lifting height is [32s, 35s]. If the time for the current lifting system to reach the same target lifting height is =29s, which is less than the lower limit of the time interval, the characteristic parameter is not activated in the performance degradation database.

[0079] The value of the first threshold value requires prior knowledge of a comprehensive good performance knowledge base and a performance degradation knowledge base to determine whether to issue an alarm signal indicating that the corresponding system needs to perform performance evaluation.

[0080] As the life cycle of the unmanned heavy-duty vehicle progresses, the production data of the unmanned heavy-duty vehicle between two adjacent maintenances will gradually stabilize. The more times the current production data exceeds the historical production data stored in the good performance knowledge base and the performance degradation knowledge base, the higher the probability of requiring maintenance intervention.

[0081] Step S5, perform a performance evaluation on the system targeted by the alarm signal. If the evaluation result is that the scrap condition has not been reached and maintenance intervention is required, the unmanned heavy-duty vehicle is classified as a second type of vehicle, and step S2 is executed to update the performance degradation database; then the system is repaired, and the unmanned heavy-duty vehicle after the performance is restored is classified as a first type of vehicle, and step S2 is executed to update the good performance knowledge base, and a maintenance record is added at the same time; if the evaluation result is that the scrap condition has not been reached and the time for maintenance intervention has not come, the characteristic parameter values ​​under typical working conditions in the good performance knowledge base are added. The upper and lower limits of the interval are taken to reduce the upper and lower limits of the interval of characteristic parameter values ​​under typical working conditions in the performance attenuation database; if the evaluation result is that the system is completely damaged, further evaluate whether it is necessary to replace the system. If necessary, replace the system and classify the unmanned heavy-duty vehicle after performance restoration as a first type of vehicle, execute step S2, and update the good performance knowledge base; if it is assessed that the unmanned heavy-duty vehicle has reached the scrap condition, exit the online monitoring of the unmanned heavy-duty vehicle and scrap the unmanned heavy-duty vehicle.

[0082] By updating the good performance database and the performance degradation database in real time, the good performance database and the performance degradation database can be adjusted in real time according to the operating status of each system of the unmanned heavy-duty vehicle, and the judgment criteria of good performance and performance degradation can be corrected, so as to more accurately judge the performance of each system.

[0083] The embodiments of the present invention are described above, but the present invention is not limited to the above-mentioned specific implementation modes. The above-mentioned specific implementation modes are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, all of which are protected by the present invention.

Claims

1. A method for online monitoring of system performance of an unmanned heavy-duty vehicle, characterized in that: The steps include: Step S1, extracting typical operating conditions of each system of an unmanned heavy-duty vehicle and the duration of each typical operating condition from historical data; Step S2, providing an unmanned heavy-duty vehicle with good system performance, which is defined as a first type of vehicle; Provide an unmanned heavy-duty vehicle that just needs maintenance intervention, which is defined as a second type of vehicle; test the system response of each system of the first type of vehicle and the second type of vehicle under typical working conditions according to different operating environments, respectively calculate the transfer function of the response of each system under different operating environments and different typical working conditions with respect to the input of the system, extract characteristic parameters of each system response transfer function, and construct a good performance knowledge base with the characteristic parameters of the system response transfer function of the first type of vehicle and production data; construct a performance degradation database with the characteristic parameters of the system response transfer function of the second type of vehicle and production data; Step S3, for online monitoring of the system performance of any unmanned heavy-duty vehicle, first obtain the operating environment in which the unmanned heavy-duty vehicle is running, then use the upper limit of the duration of each typical working condition extracted in step S1 as the length of the sliding time window, monitor the system input and system response of each system of the unmanned heavy-duty vehicle online, and use the mutual correlation coefficient method to match the system input of any system of the unmanned heavy-duty vehicle with the system input under the typical working condition. If the match is successful, save the system input and system response collected in the current sliding time window, use the least squares method to fit the transfer function of the system response with respect to the system input, and extract the characteristic parameters of the transfer function; Step S4, using the least squares method to compare the characteristic parameters obtained in step S3 with the characteristic parameters corresponding to each system under typical working conditions matched in the good performance knowledge base. If the characteristic parameters obtained in step S3 are greater than the upper limit of the characteristic parameter value range under typical working conditions in the good performance knowledge base, the characteristic parameters are deemed to be activated; at the same time, the characteristic parameters corresponding to each system under typical working conditions matched in the performance decay database are compared. If the characteristic parameters obtained in step S3 are greater than the lower limit of the characteristic parameter value range under typical working conditions in the performance decay database, the characteristic parameters are deemed to be activated; the number of activated characteristic parameters is counted, and if the number of activated characteristic parameters exceeds a preset first threshold, an alarm signal is issued that the corresponding system needs to perform performance evaluation; in addition, the current production data is compared with the production data stored in the good performance knowledge base and the performance decay knowledge base, and the total number of times the current production data is greater than the production data in the good performance knowledge base and the performance decay knowledge base is counted. If the total number is greater than a preset second threshold, an alarm signal is also issued that the corresponding system needs to perform performance evaluation; Step S5, perform a performance evaluation on the system targeted by the alarm signal. If the evaluation result is that the scrap condition has not been reached and maintenance intervention is required, the unmanned heavy-duty vehicle is classified as a second type of vehicle, and step S2 is executed to update the performance degradation database; then the system is repaired, and the unmanned heavy-duty vehicle after the performance is restored is classified as a first type of vehicle, and step S2 is executed to update the good performance knowledge base, and a maintenance record is added at the same time; if the evaluation result is that the scrap condition has not been reached and the time for maintenance intervention has not come, the characteristic parameter values ​​under typical working conditions in the good performance knowledge base are added. The upper and lower limits of the interval are taken to reduce the upper and lower limits of the interval of characteristic parameter values ​​under typical working conditions in the performance attenuation database; if the evaluation result is that the system is completely damaged, further evaluate whether it is necessary to replace the system. If necessary, replace the system and classify the unmanned heavy-duty vehicle after performance restoration as a first type of vehicle, execute step S2, and update the good performance knowledge base; if it is assessed that the unmanned heavy-duty vehicle has reached the scrap condition, exit the online monitoring of the unmanned heavy-duty vehicle and scrap the unmanned heavy-duty vehicle.

2. The method for online monitoring of unmanned heavy-duty vehicle system performance according to claim 1 is characterized in that: The unmanned heavy-load vehicle is an unmanned heavy-load vehicle for mining, and the monitoring system includes a steering system in normal mode, a braking system, a drive system in normal mode, and a lifting system, wherein: typical operating conditions of the steering system include obstacle avoidance conditions, U-turn conditions, and right-angle turning conditions at various vehicle speeds; typical operating conditions of the braking system include different brake pedal opening conditions at various initial vehicle speeds under straight roads, downhill roads, and uphill roads, and under different load conditions; typical operating conditions of the drive system include different accelerator pedal opening conditions at various initial vehicle speeds under straight roads, downhill roads, and uphill roads, and under different load conditions; typical operating conditions of the lifting system include conditions of lifting cargo boxes of different load masses according to different lifting times.

3. The method for online monitoring of unmanned heavy-duty vehicle system performance according to claim 1 is characterized in that: The first type of vehicles includes brand new unmanned heavy-duty vehicles and unmanned heavy-duty mining vehicles whose performance has been restored after repair.

4. The method for online monitoring of unmanned heavy-duty vehicle system performance according to claim 1 is characterized in that: The operating environment includes weather, temperature and road environment, wherein the road environment includes dry road, wet road and frozen road.

5. The method for online monitoring of unmanned heavy-duty vehicle system performance according to claim 2 is characterized in that: The system input of the steering system is the steering wheel angle; the system response of the steering system includes yaw rate, lateral acceleration, steering angle, steering system oil pressure and steering hydraulic lever stroke; the system input of the braking system is the brake pedal opening; the system response of the braking system includes longitudinal acceleration, longitudinal speed, brake hydraulic cylinder pressure and wheel speed; the system input of the drive system is the accelerator pedal opening; The system response of the driving system includes longitudinal acceleration, longitudinal speed, driving motor torque and wheel speed; the system input of the lifting system is lifting time; the system response of the lifting system includes bucket lifting angle, lifting hydraulic cylinder pressure and lifting hydraulic cylinder stroke.

6. The method for online monitoring of unmanned heavy-duty vehicle system performance according to claim 5 is characterized in that: The transfer functions of the steering system and the drive system are fitted using an underdamped second-order system transfer function model. The underdamped second-order system transfer function is expressed as: ； Where s represents the complex frequency variable in Laplace transform; represents the natural frequency of the system, represents the system damping ratio and satisfies ; The transfer functions of the braking system and the lifting system are fitted using a first-order inertial system transfer function model. The first-order inertial system transfer function is expressed as: ； In the formula, Represents the system time constant.

7. The method for online monitoring of unmanned heavy-duty vehicle system performance according to claim 6 is characterized in that: The characteristic parameters of the transfer function of the steering system and the drive system include: damped oscillation frequency , Rise time , Peak Time , Overshoot , Adjust time As well as the system damping ratio, the definitions of the characteristic parameters are as follows: ； ； ； ； ； The characteristic parameters of the transfer function of the braking system and the lifting system include: system time constant , first-order system rise time And the first-order system adjustment time , where the first-order system rise time is It represents the time for the system to go from 10% to 90% in step response. ; First-order system adjustment time .

8. The method for online monitoring of unmanned heavy-duty vehicle system performance according to claim 1, characterized in that: The good performance knowledge base and the performance degradation knowledge base have the same data structure, both adopting a tree structure, each system corresponds to a first-level branch, each first-level branch is provided with at least one second-level branch, each second-level branch corresponds to a maintenance state, and the data before maintenance is stored in the performance degradation knowledge base; the data after maintenance is stored in the good performance knowledge base; each second-level branch is provided with at least one third-level branch, each third-level branch corresponds to a road surface environment; each third-level branch is provided with at least one fourth-level branch, each fourth-level branch corresponds to a typical working condition; each fourth-level branch is provided with at least one fifth-level branch, each fifth-level branch corresponds to a characteristic parameter of a system response transfer function.

9. The method for online monitoring of unmanned heavy-duty vehicle system performance according to claim 1, characterized in that: The new state of the unmanned heavy-duty vehicle is taken as the initial node, and each maintenance state corresponds to a node. The production data includes the total mass of different types of goods transported by the unmanned heavy-duty vehicle between two adjacent nodes, the corresponding mileage, and the accumulated work done in the vertical direction to overcome gravity; in the new state, the production data of the unmanned heavy-duty vehicle is 0.

10. An unmanned heavy-duty vehicle system performance online monitoring system, characterized in that: Used to execute the unmanned heavy-duty vehicle system performance online monitoring method described in any one of claims 1-9.

Images