New energy vehicle predictive maintenance method and system based on cloud data analysis

Abstract

The application relates to the technical field of vehicle maintenance, in particular to a new energy vehicle predictive maintenance method and system based on cloud data analysis. The method comprises the following steps: acquiring a vehicle operation dynamic information set, analyzing the space-time correlation between brake disc degradation and bus line characteristics based on the vehicle operation dynamic information set, and obtaining predictive maintenance guide information; based on the predictive maintenance guide information, tracking vehicle operation behavior data, calculating a real-time cumulative load spectrum of a bus line, triggering a vehicle personalized maintenance behavior, and obtaining a vehicle maintenance information set; based on the vehicle maintenance information set, dynamically adjusting a bus maintenance control strategy, and outputting a new energy vehicle predictive maintenance log. Dynamically adjusting the maintenance strategy adapts to the actual state of the vehicle, reduces the new energy bus operation interruption caused by faults, and improves the intelligent level and comprehensive operation benefit of the new energy bus maintenance.

Description

Technical Field

[0001] This application relates to the field of vehicle maintenance technology, and in particular to a predictive maintenance method and system for new energy vehicles based on cloud data analysis. Background Technology

[0002] Currently, the maintenance of new energy buses mainly adopts regular planned maintenance and real-time alarm mechanisms based on simple thresholds. Existing solutions usually conduct health assessments and alarms based on overall fleet data or general models, which fail to be deeply integrated with the dynamic operating characteristics of specific bus routes.

[0003] However, in real-world scenarios where bus routes are fixed but road conditions and passenger flow are complex and ever-changing, the above methods have systemic limitations: they fail to establish targeted degradation models based on the unique gradient, congestion, and passenger flow changes of each route; the fusion of multi-source dynamic data is insufficient, making it difficult to accurately characterize the impact of both regular and sudden loads on component fatigue accumulation; and maintenance strategies lack adaptive adjustment capabilities based on route clustering and real-time environment, resulting in delayed early warnings and low maintenance efficiency. Currently, in the field of new energy bus operation and maintenance, there is still a lack of a complete technical system for accurately predicting and personalized maintenance of brake disc thermal fatigue and cracks through cloud data analysis and the integration of real-time multi-dimensional data. Summary of the Invention

[0004] This application provides a predictive maintenance method and system for new energy vehicles based on cloud data analysis to solve the above problems.

[0005] In a first aspect, this application provides a predictive maintenance method for new energy vehicles based on cloud data analysis. The method includes: acquiring a dynamic information set of vehicle operation; analyzing the spatial-temporal correlation between brake disc degradation and bus route characteristics based on the dynamic information set of vehicle operation to obtain predictive maintenance guidance information; tracking vehicle operation behavior data based on the predictive maintenance guidance information, calculating the real-time cumulative load spectrum of the bus route, triggering personalized maintenance behavior of the vehicle, and obtaining a vehicle maintenance information set; and dynamically adjusting the bus maintenance control strategy based on the vehicle maintenance information set to output a predictive maintenance log for new energy vehicles.

[0006] Through the above technical solutions, the predictive maintenance method uses cloud data analysis to establish a spatial-temporal correlation between brake disc degradation and line characteristics, enabling accurate prediction of maintenance needs. By adopting personalized maintenance behaviors, it avoids over-maintenance and under-maintenance, reducing operating costs. It also dynamically adjusts maintenance strategies to adapt to the actual condition of the vehicle, extending the life of key components, ensuring driving safety, reducing operational interruptions caused by faults, and improving the level of intelligent maintenance and overall operational efficiency of new energy buses.

[0007] Optionally, the step of analyzing the spatial-temporal correlation between brake disc degradation and bus route characteristics based on the vehicle operation dynamic information set to obtain predictive maintenance guidance information includes: the vehicle operation dynamic information set includes historical vehicle braking data, route geographic information, and route operating timetable; based on the route geographic information, analyzing the differentiated impact of spatial characteristics of each segment of the bus operating route on brake disc thermal load to obtain brake disc spatial degradation information; based on the vehicle historical braking data and combined with the route operating timetable, analyzing the temporal variation law of braking behavior of the vehicle in different operating segments of the same route to obtain brake disc temporal degradation information; integrating the brake disc spatial degradation information and the brake disc temporal degradation information to construct a spatial-temporal correlation rule reflecting the coupling between route characteristics and brake disc degradation to obtain the predictive maintenance guidance information.

[0008] Optionally, the process of constructing the brake disc spatial degradation information includes: identifying and extracting long downhill sections, continuous curved sections, and densely packed platform sections in the line based on the line's geographical information as key sections for braking thermal load; analyzing the spatial synergistic effect and thermal accumulation effect of braking behavior between adjacent key sections based on the key sections for braking thermal load to obtain braking load correlation information between sections; and analyzing the line spatial degradation characteristics reflecting the differences in the contribution of different spatial locations to the brake disc thermal load based on the key sections for braking thermal load and the braking load correlation information between sections to obtain the brake disc spatial degradation information.

[0009] Optionally, the process of constructing the brake disc time-series degradation information includes: based on the route operating timetable and combined with the vehicle's historical braking data, analyzing the distribution characteristics of bus braking frequency and braking intensity during each period of morning peak, evening peak, off-peak, and nighttime low-peak to obtain time-series braking information; based on the time-series braking information, analyzing the differentiated effects of frequent starts and stops and high-intensity braking during peak hours, medium-intensity braking during stable operation during off-peak hours, and low-intensity braking during low-peak hours on brake disc thermal fatigue accumulation to obtain time-series braking load characteristics; based on the time-series braking information and the time-series braking load characteristics, analyzing the degradation time-series curve reflecting the change of brake disc degradation rate with time band to obtain the brake disc time-series degradation information.

[0010] Optionally, the analysis of the degradation time-series curve reflecting the change of brake disc degradation rate with time bands includes: based on the time-series braking load characteristics, analyzing the baseline level of brake disc degradation rate in each period of the morning peak, the evening peak, the off-peak period, and the nighttime low-peak period, as well as the step change law of the baseline level as the operating period switches, to obtain the degradation rate time-series baseline spectrum; based on the degradation rate time-series baseline spectrum, combined with the time-series braking information, analyzing the inheritance effect of accumulated thermal fatigue in the previous operating period on the initial degradation state in the next period, as well as the dynamic modulation effect of the inheritance effect on the degradation rate in the next period, to obtain the inter-period degradation rate modulation relationship; and constructing the degradation time-series curve that fully reflects the continuous and adaptive change of brake disc degradation rate in different time bands based on the degradation rate time-series baseline spectrum and the inter-period degradation rate modulation relationship.

[0011] Optionally, the process of constructing the spatial-temporal association rule includes: based on the brake disc spatial degradation information and combined with the brake disc temporal degradation information, extracting the degradation performance of the key road segments of the braking heat load during the morning peak, the evening peak, the off-peak period, and the nighttime low peak period, to obtain a key road segment-time period degradation feature set with spatiotemporal labels; based on the key road segment-time period degradation feature set, identifying the coupling phenomenon where the degradation contribution of different spatial road segments is amplified or suppressed under different time periods, to obtain the joint modulation relationship between spatial features and time period features on the degradation rate; based on the joint modulation relationship, dynamically associating and mapping the line spatial degradation features with the degradation time series curve to generate a composite correlation degree used to quantify the strength of brake disc degradation risk at any operating time and any line position, and using the calculation and matching logic of the composite correlation degree as the spatial-temporal association rule; the line spatial degradation features and the degradation time series curve are bidirectionally coupled and positively correlated.

[0012] Optionally, the process of constructing the vehicle maintenance information set includes: according to the spatial-temporal association rules, real-time vehicle location data and time data are obtained, and the composite correlation degree corresponding to the current operating time and the location of the bus route is matched to obtain the real-time composite correlation degree; based on the real-time composite correlation degree, the comprehensive degradation state of the current brake disc caused by the combined effect of real-time operating load and historical accumulated degradation information is analyzed to obtain the real-time accumulated load spectrum; based on the real-time accumulated load spectrum, maintenance warnings and execution instructions are triggered according to the current comprehensive degradation state, and maintenance levels, suggested maintenance time windows and maintenance operation items are generated for the corresponding bus vehicles to obtain the vehicle maintenance information set.

[0013] Optionally, the process of constructing the real-time cumulative load spectrum includes: based on the real-time composite correlation degree, analyzing the real-time operating load intensity level corresponding to the current operating time and the location of the bus route to obtain the real-time load level; based on the historical cumulative degradation information, analyzing the instantaneous load superposition effect and dynamic acceleration effect of the real-time load level on the historical cumulative degradation information to obtain the real-time load effect characteristics; based on the real-time load effect characteristics, dynamically correcting the historical cumulative degradation information to generate a comprehensive degradation state assessment value that integrates the impact of the current real-time operating load and the historical degradation accumulation results; and using the set of the comprehensive degradation state assessment value continuously updated with the operation of the bus to affect the brake disc as the bus runs as the real-time cumulative load spectrum.

[0014] Optionally, the step of dynamically adjusting the bus maintenance control strategy based on the vehicle maintenance information set and outputting predictive maintenance logs for new energy vehicles includes: based on the vehicle maintenance information set, analyzing the aggregation trend of maintenance needs and the timing of collaborative maintenance among multiple vehicles according to the real-time cumulative load spectrum of different vehicles on different routes, and obtaining a cluster maintenance scheduling strategy; based on the cluster maintenance scheduling strategy, dynamically optimizing and merging the suggested maintenance time windows of each vehicle to generate a collaborative maintenance execution sequence that takes into account both the urgency of individual vehicle degradation and the overall maintenance resource efficiency; scheduling and executing specific maintenance operations according to the collaborative maintenance execution sequence, and automatically recording the actual load characteristics, maintenance measures, and effect verification data during the maintenance process, integrating them into a cloud database to form a structured predictive maintenance log for new energy vehicles.

[0015] Secondly, this application provides a predictive maintenance system for new energy vehicles based on cloud data analysis, the system comprising:

[0016] The maintenance guidance analysis module is used to acquire a set of dynamic vehicle operation information. Based on this set, it analyzes the spatial-temporal correlation between brake disc degradation and bus route characteristics to obtain predictive maintenance guidance information. The cumulative load analysis module is used to track vehicle operation behavior data based on the predictive maintenance guidance information, calculate the real-time cumulative load spectrum of the bus route, trigger personalized vehicle maintenance behaviors, and obtain a set of vehicle maintenance information. The maintenance control analysis module is used to dynamically adjust the bus maintenance control strategy based on the set of vehicle maintenance information and output predictive maintenance logs for new energy vehicles. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram illustrating an application scenario provided in one embodiment of this application;

[0019] Figure 2 A flowchart illustrating a predictive maintenance method for new energy vehicles based on cloud data analysis, provided as an embodiment of this application;

[0020] Figure 3 This is a schematic diagram of the structure of a predictive maintenance system for new energy vehicles based on cloud data analysis, provided as an embodiment of this application. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0022] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0023] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.

[0024] In the daily operation of new energy buses, bus routes are fixed but road conditions and passenger flow are complex and changeable. Existing methods do not establish degradation models for changes in route gradient, congestion, and passenger flow; the integration of multi-source dynamic data is insufficient, making it difficult to accurately characterize the impact of load on component fatigue accumulation; maintenance strategies lack adaptive adjustment capabilities, resulting in delayed early warnings and low efficiency. The field of new energy bus operation and maintenance still lacks a complete technical system that can accurately predict brake disc thermal fatigue and cracks and perform personalized maintenance through cloud and multi-dimensional real-time data.

[0025] Based on this, this application provides a predictive maintenance method and system for new energy vehicles based on cloud data analysis. The predictive maintenance method relies on cloud data analysis to construct a spatiotemporal correlation model of brake disc degradation and line characteristics, accurately predict maintenance needs, avoid over-maintenance and under-maintenance problems through personalized maintenance, and dynamically adapt strategies to match the actual condition of the vehicle. This not only reduces operating costs and extends component life, but also ensures driving safety and reduces downtime due to malfunctions, thus comprehensively improving the intelligence and overall efficiency of new energy bus maintenance.

[0026] Figure 1 This application provides an illustration of an application scenario. In the daily operation of new energy buses, the method provided in this application is applied to construct the spatiotemporal correlation between brake disc degradation and line characteristics based on cloud data analysis. By accurately predicting maintenance needs and dynamically adjusting maintenance strategies, the maintenance of new energy buses can be made intelligent, reducing operating costs, ensuring driving safety, and improving overall efficiency.

[0027] Specifically, the method provided in this application can be applied to any server. The server interacts with the on-board sensors to obtain the vehicle operation dynamic information set provided by the on-board sensors, realizes accurate prediction of maintenance needs, generates a vehicle maintenance information set for the driver through personalized maintenance behavior, ensures driving safety, and outputs predictive maintenance logs of new energy vehicles to the bus maintenance and operation platform, thereby improving the intelligent level of new energy bus maintenance and the overall operational efficiency.

[0028] For specific implementation details, please refer to the following examples.

[0029] Figure 2 This is a flowchart illustrating a predictive maintenance method for new energy vehicles based on cloud data analysis, provided as an embodiment of this application. The method of this embodiment can be applied to the server in the above scenario. Figure 2 As shown, the method includes:

[0030] S201. Obtain the vehicle operation dynamic information set. Based on the vehicle operation dynamic information set, analyze the spatial-temporal correlation between brake disc degradation and bus route characteristics to obtain predictive maintenance guidance information.

[0031] The vehicle operation dynamic information set can be a collection reflecting various dynamic states during the operation of new energy buses, with onboard sensors used as the data source. Brake disc degradation refers to the performance decline of brake discs in new energy buses due to friction wear and stress during long-term use. Bus route characteristics can be the inherent attributes and operating environment characteristics of the new energy bus operating routes. Spatial-temporal correlation can be the intrinsic relationship between brake disc degradation status and bus route characteristics in spatial distribution and time dimensions. Predictive maintenance guidance information can be forward-looking information derived from the analysis of the vehicle operation dynamic information set, used to guide subsequent maintenance work.

[0032] Specifically, the reliability of the braking system of new energy buses is directly related to driving safety and operational efficiency. Existing maintenance methods are based on fixed mileage or time intervals and lack accurate assessment of the actual operating status of the vehicle, which may lead to over-maintenance or under-maintenance. Brake disc degradation is significantly affected by road characteristics. For example, frequent braking sections or steep slopes will accelerate wear. By analyzing spatial-temporal correlations, the life of brake discs can be predicted more accurately, enabling predictive maintenance, reducing unexpected failures and lowering maintenance costs.

[0033] S202. Based on predictive maintenance guidance information, track vehicle operation behavior data, calculate the real-time cumulative load spectrum of bus routes, trigger personalized vehicle maintenance behavior, and obtain a vehicle maintenance information set.

[0034] Vehicle operation behavior data can be the specific driving operations and status data of a vehicle during actual operation. Real-time cumulative load spectrum of a bus route can be a comprehensive data spectrum reflecting the load pressure exerted by the route on components such as the vehicle's braking system, calculated cumulatively based on vehicle operation behavior data. Personalized maintenance behavior can be targeted maintenance operations formulated based on personalized data such as the load characteristics of the route to which the vehicle belongs, and the trend of brake disc degradation. Vehicle maintenance information set can be a collection of relevant data recording the entire process of personalized vehicle maintenance behavior.

[0035] Specifically, even on the same route, the operating behavior of different vehicles may vary due to factors such as driver habits and passenger load, resulting in different cumulative loads. Existing maintenance strategies often ignore these individual differences and adopt a one-size-fits-all approach, which can easily lead to resource waste or insufficient maintenance risks. By calculating the real-time cumulative load spectrum, the actual load of the vehicle can be quantified. Combined with predictive maintenance guidance information, personalized maintenance behavior can be triggered to ensure that the maintenance timing matches the actual condition of the vehicle.

[0036] S203. Based on the vehicle maintenance information set, dynamically adjust the bus maintenance control strategy and output the predictive maintenance log of new energy vehicles.

[0037] Bus maintenance control strategies can be rules and plans that guide the maintenance work of new energy buses. Predictive maintenance logs for new energy vehicles can be documents that record key information throughout the entire predictive maintenance process.

[0038] Specifically, maintenance control strategies need to be continuously optimized as vehicle maintenance history accumulates to adapt to changes in the overall fleet status. Static strategies are difficult to cope with dynamic factors such as vehicle aging and route adjustments. By dynamically adjusting strategies based on maintenance information sets, maintenance resources can be rationally allocated, improving the overall availability of the fleet.

[0039] The predictive maintenance method provided in this embodiment uses cloud data analysis to establish a spatial-temporal correlation between brake disc degradation and line characteristics, enabling accurate prediction of maintenance needs. Through personalized maintenance behavior, it avoids over-maintenance and under-maintenance, reducing operating costs; dynamically adjusts maintenance strategies to adapt to the actual condition of the vehicle, extends the life of key components, ensures driving safety, reduces operational interruptions caused by faults, and improves the level of intelligent maintenance and overall operational efficiency of new energy buses.

[0040] In some embodiments, the vehicle operation dynamic information set includes historical vehicle braking data, route geographic information, and route operating timetables. Based on the route geographic information, the spatial characteristics of each segment of the bus operating route are analyzed to determine the differential impact on the brake disc thermal load, thus obtaining brake disc spatial degradation information. Based on the historical vehicle braking data and combined with the route operating timetables, the temporal variation pattern of the vehicle's braking behavior in different operating segments on the same route is analyzed, thus obtaining brake disc temporal degradation information. The brake disc spatial degradation information and brake disc temporal degradation information are integrated to construct a spatial-temporal correlation rule that reflects the coupling between route characteristics and brake disc degradation, thereby obtaining predictive maintenance guidance information.

[0041] Historical braking data can be various records related to braking operations generated during the past operation of buses. Route geographic information can be a collection of information that digitally describes the geographic features of the areas traversed by the bus routes. Route timetables can be information that specifies the planned departure times of buses on each operating route, the estimated arrival and departure times of each stop, and the departure intervals at different times. Brake disc spatial degradation information can be information on the differences in the contribution of different route segments to the thermal load of the brake discs. Brake disc temporal degradation information can be information on how the brake disc degradation rate changes with different operating periods.

[0042] Specifically, in the daily operation of new energy buses, brake disc degradation is closely related to the spatial characteristics of the route and the operating status during different time periods. Current maintenance neglects the differentiated thermal load effects of road sections (such as steep slopes and continuous curves) and time periods (such as frequent starts and stops during peak hours), which can easily lead to brake failure and traffic accidents due to untimely maintenance, or excessive maintenance increasing operating costs and shortening the service life of the brake discs. To address these issues: First, spatial analysis technology using Geographic Information Systems (GIS) is used to process the input route geographic information. For example, slope analysis algorithms can automatically identify long downhill sections with a gradient consistently greater than 5%, and curvature calculations can be used to delineate continuous curve sections with a turning radius of less than 50 meters. Combined with station coordinate density analysis, densely populated station areas are identified, and these sections are collectively defined as critical sections for brake thermal load. Second, based on historical vehicle braking data (such as brake pressure, ... By combining frequency time series data with route timetables, and employing temporal clustering and statistical analysis, the daily operation is divided into typical time periods such as morning peak (e.g., 7:00-9:00), evening peak (e.g., 17:00-19:00), off-peak period, and nighttime low-peak period. The average braking frequency (e.g., the morning peak can be twice that of the off-peak period) and intensity distribution within each time period are quantitatively analyzed. Furthermore, through data fusion and association rule mining, the spatial and temporal analysis results are dynamically coupled. For example, by analyzing historical braking data under the spatiotemporal combination of "vehicles traveling on a specific long downhill section during the morning peak period," the phenomenon that the brake disc temperature rise rate and wear rate are significantly higher than "the same section during the off-peak period" is quantified. This allows the construction of spatial-temporal association rules that can quantify the strength of brake disc degradation risk corresponding to any "time-location" point, ultimately forming predictive maintenance guidance information that can guide subsequent differentiated maintenance.

[0043] The method provided in this embodiment accurately captures the degradation patterns of different road sections and time periods, providing a scientific basis for subsequent maintenance. This not only avoids the waste of resources caused by blind maintenance, but also provides early warning of potential faults, reduces the risk of brake failure, and ensures driving safety.

[0044] In some embodiments, based on the geographical information of the route, long downhill sections, continuous curved sections, and sections with dense platform areas are identified and extracted as key sections for braking thermal load. Based on the key sections for braking thermal load, the spatial synergistic effect and thermal accumulation effect of braking behavior between adjacent key sections are analyzed to obtain braking load correlation information between sections. Based on the key sections for braking thermal load and the braking load correlation information between sections, the spatial degradation characteristics of the route reflecting the differences in the contribution of different spatial locations to the thermal load of the brake disc are analyzed to obtain spatial degradation information of the brake disc.

[0045] Critical sections of braking thermal load can be sections of a railway line where frequent and intense braking occurs due to road or operational characteristics, resulting in significant thermal load impacts on the brake discs. Spatial synergy effects refer to the mutual influence and correlation of braking behaviors between adjacent critical sections of braking thermal load. Thermal accumulation effects refer to the phenomenon where the thermal load generated by braking behaviors in adjacent critical sections of braking thermal load is not dissipated in a timely manner and continues to accumulate. Inter-segment braking load correlation information can record the spatial synergy and thermal accumulation effects of braking behaviors between adjacent critical sections of braking thermal load. Line spatial degradation characteristics can be the differentiated attributes of road segments in different spatial locations regarding the degree of contribution and accumulation mode of braking thermal load.

[0046] Specifically, during bus operation, failure to construct brake disc spatial degradation information can lead to inaccurate brake disc degradation analysis due to neglecting differences in route geographical features. Braking loads on long downhill slopes and continuous curves, as well as the heat accumulation effect between road sections, are not considered. This can result in delayed maintenance of high-risk sections, causing safety hazards, while excessive maintenance of low-risk sections increases costs. To address these issues, the following approach is employed: First, geographic information parsing and feature rule matching technologies are used to process the route geographic information provided in the cloud. For example, by calculating the elevation difference and horizontal distance between three consecutive points in the digital elevation model (DEM) data, the slope duration can be automatically identified. Sections exceeding a set threshold (e.g., 4%) and length exceeding a specific distance (e.g., 800 meters) are designated as "long downhill sections." By calculating the turning angle sequence and radius of curvature of the route trajectory, areas with multiple consecutive curves having radii less than the standard value (e.g., 60 meters) are identified as "continuous curve sections." Through spatial overlay analysis with a bus stop database, areas with significantly lower-than-average station spacing (e.g., more than three consecutive stations within 300 meters) are identified as "dense platform areas." This process completes the extraction and labeling of all key road sections. Subsequently, path analysis and heat analysis based on a physical model are employed. Mechanical simulation methods are used to analyze road segment correlations. For example, a dynamic model incorporating vehicle mass, speed, and braking parameters is constructed to simulate the continuous process of a vehicle entering a sharp curve (critical segment B) from the end of a long downhill section (critical segment A) with a high residual speed, calculating the energy conversion between the two braking actions. Furthermore, a lumped-parameter thermal model is used to calculate the residual temperature of the disc brakes generated by braking in segment A at the beginning of segment B. This quantifies the degree to which the heat accumulation in segment A improves the initial thermal state of segment B, thus forming a "correlation matrix" describing the intensity of heat load transfer between any two critical segments. Finally, Degradation features are constructed by applying spatial feature fusion and dynamic weight mapping. For example, a basic load index is assigned to each key road segment according to its type and intensity (such as slope value and curve sharpness). Then, the index of downstream road segments with strong heat transfer relationship is dynamically added according to the aforementioned correlation matrix (such as increasing the index of curved road segments that are immediately followed by long downhill slopes by 30%). Finally, a "road spatial degradation feature" vector is generated, which is indexed by the line station number or location ID and corresponds to a comprehensive spatial load score for each point. This vector is the structured brake disc spatial degradation information, which can be directly input into the subsequent spatiotemporal correlation model.

[0047] The method provided in this embodiment accurately identifies key road sections with braking thermal load and the correlation effects between road sections, clarifies the differences in degradation contribution at different spatial locations, makes brake disc degradation analysis more consistent with actual road scenarios, provides accurate spatial data support for the subsequent construction of spatial-temporal correlation rules, improves the pertinence and scientific nature of braking system maintenance, and extends the service life of brake discs.

[0048] In some embodiments, based on the route operating timetable and combined with historical vehicle braking data, the distribution characteristics of bus braking frequency and braking intensity during each period of morning peak, evening peak, off-peak, and nighttime low-peak are analyzed to obtain time-specific braking information. Based on the time-specific braking information, the differentiated effects of frequent starts and stops and high-intensity braking during peak hours, medium-intensity braking during stable operation during off-peak hours, and low-intensity braking during low-peak hours on brake disc thermal fatigue accumulation are analyzed to obtain time-sequential braking load characteristics. Based on the time-specific braking information and time-sequential braking load characteristics, the degradation time-sequential curve reflecting the change of brake disc degradation rate with time bands is analyzed to obtain brake disc time-sequential degradation information.

[0049] Morning rush hour can be a period of concentrated urban commuting demand and heavy traffic (e.g., 7:00-9:00). Evening rush hour can be a period of concentrated return commuting demand and heavy traffic (e.g., 17:00-19:00). Off-peak hours can be the period between morning and evening rush hours, when commuting demand is low and traffic is smooth (e.g., 10:00-16:00). Nighttime low-peak hours can be a period of reduced urban activity and lower public transport frequency (e.g., 23:00-06:00 the next day). Braking frequency can be the number of times a bus brakes per unit of time. Braking intensity can be the force applied when a bus brakes. Time-specific braking information can be an integration of the distribution characteristics of braking frequency and intensity across different time periods. Time-sequential braking load characteristics can be the core characteristics that differentiate the impact of braking behavior at different time periods on the thermal fatigue accumulation of the brake disc. Degradation time-sequential curves can be curves showing the dynamic changes in the brake disc degradation rate over different time segments.

[0050] Specifically, in the daily operation of public transportation, ignoring the differences in braking behavior at different times can lead to inaccurate brake disc maintenance: high-frequency, high-intensity braking during morning and evening peak hours causes rapid accumulation of thermal fatigue, easily leading to brake failure risks; excessive maintenance during off-peak and low-peak periods wastes resources. To address these issues: First, data mining and statistical analysis are used. Based on the time frame defined by the bus company's route timetable (e.g., classifying weekdays from 7:00-9:00 as the morning peak), massive amounts of historical vehicle braking data stored in the cloud (including timestamps and deceleration values ​​for each braking action) are sliced ​​and aggregated for time periods. This quantifies and extracts the distribution characteristics of braking frequency and intensity within each time period (e.g., calculating an average of 5.2 braking events per kilometer during the morning peak, with 35% of these events exceeding 0.5g in intensity), forming structured time-specific braking information. Next, a segmented analysis method based on operating time periods is applied, combined with thermodynamics and material fatigue analysis principles, to deeply analyze these characteristics: for frequent starts and stops and high-intensity braking during peak hours, the rapid temperature fluctuations on the brake disc surface caused in a short period are analyzed. The analysis examines the accelerating effect of braking on thermal fatigue accumulation. For medium-intensity braking during off-peak hours, it assesses the linear contribution of the continuous, uniform thermal load to material wear. For low-intensity braking during off-peak hours, it evaluates the near-resting state. This analysis transforms qualitative models into quantitative indicators, yielding temporal braking load characteristics. Finally, using nonlinear degradation modeling methods (such as curve fitting based on piecewise functions or state-space models), with time-specific braking information as input and temporal braking load characteristics as modulation parameters, a continuous degradation time-series curve is dynamically simulated and generated. This curve visually demonstrates the complete band change process of the degradation rate: a sharp increase during the morning peak, a slow increase during the midday off-peak, a renewed surge during the evening peak, and a gradual leveling off at night. This curve and its characteristic parameters constitute the final required temporal degradation information for the brake disc.

[0051] The method provided in this embodiment accurately captures the differentiated impact of braking on brake disc degradation at different time periods, clearly presents the degradation time sequence pattern, provides key time sequence data for the construction of spatial-temporal correlation rules, makes maintenance guidance information more targeted, avoids untimely or excessive maintenance, optimizes resource allocation, reduces operating costs, and extends brake disc life.

[0052] In some embodiments, based on the time-series braking load characteristics, the baseline levels of brake disc degradation rates during each period of morning peak, evening peak, off-peak, and nighttime low-peak are analyzed, as well as the step change law of the baseline levels as the operating period switches, to obtain the time-series baseline spectrum of degradation rates. Based on the time-series baseline spectrum of degradation rates, combined with time-series braking information, the inheritance effect of accumulated thermal fatigue in the previous operating period on the initial degradation state in the next period, and the dynamic modulation effect of the inheritance effect on the degradation rate in the next period are analyzed, to obtain the time-series degradation rate modulation relationship. Based on the time-series baseline spectrum of degradation rates and the time-series degradation rate modulation relationship, a complete degradation time-series curve reflecting the continuous and adaptive changes of brake disc degradation rate in different time bands is constructed.

[0053] The degradation rate time-period reference spectrum can be a set of reference levels for brake disc degradation rates within each time period, determined based on the braking load characteristics of different time periods, and the step-like variation patterns of these reference levels as the operating period switches. The inheritance effect can be the continued influence of accumulated thermal fatigue from braking in the previous operating period on the initial degradation state of the brake disc in the subsequent period. The dynamic modulation effect can be the dynamic adjustment effect of the inherited effect causing the brake disc degradation rate in the subsequent period to deviate from its reference level. The reference level can be the standard value of the degradation rate of the brake disc based on the braking frequency and intensity of a certain time period under ideal conditions without the influence of other time periods. The step-like variation pattern can be the discontinuous, abrupt change characteristics of the brake disc degradation rate reference level when switching between different time periods.

[0054] Specifically, in the operation of new energy buses, if the time-based baseline level, step changes, and inter-segment inheritance and modulation effects of brake disc degradation rate are ignored, it will lead to distorted judgment of degradation status. This can result in misjudgment of maintenance timing, excessive maintenance increasing operating costs, or delayed maintenance. In severe cases, excessive brake disc degradation can lead to braking performance failure in scenarios such as peak congestion and long downhill slopes, causing safety accidents. To address these issues, firstly, based on the "time-sequential braking load characteristics" data obtained from the cloud, statistical modeling and pattern recognition technologies are used to quantify and calibrate the "baseline level" of brake disc degradation rate for four typical time periods: morning peak, evening peak, off-peak, and nighttime low-peak. By analyzing a large number of time-switching instances in historical data, the pattern of baseline value transitions is extracted (for example, cluster analysis shows that the baseline value during the evening peak is generally about one level higher than that during the off-peak), thus solidifying it into a "time-based baseline spectrum of degradation rate". Subsequently, for the analysis of the "inheritance effect," data fitting and state-space modeling methods were used to specifically quantify the increase in the initial degradation state of the subsequent period by the equivalent heat load accumulated in the previous operating period (such as converting the intensity and frequency of frequent braking during the morning rush hour into a heat accumulation index). For example, it was quantified that a severe morning rush hour can increase the initial degradation rate of the subsequent off-peak period by approximately 15% above its baseline value. For the "dynamic modulation effect," a regression model was established to analyze the influence coefficient of the initial state offset on the degradation rate curve of the entire subsequent period. Finally, model fusion technology was used to integrate the time frame provided by the reference spectrum with the dynamic correction rules provided by the modulation relationship. For example, weighted superposition or state transition equations were used to generate a continuous "degradation time series curve" that can smoothly transition and whose degradation rate value at each time point integrates the current time period reference and the historical accumulated modulation effects.

[0055] The method provided in this embodiment accurately captures the degradation patterns of brake discs at different times and the mutual influence between time periods, providing accurate time-series data support for subsequent spatial-temporal correlation rules. This makes the quantification of brake disc degradation risk more consistent with actual operation, avoids untimely or excessive maintenance, ensures stable operation of the braking system, and reduces vehicle operation and maintenance costs.

[0056] In some embodiments, based on the spatial degradation information of the brake disc and combined with the temporal degradation information of the brake disc, the degradation performance of key road sections under braking heat load during morning peak, evening peak, off-peak, and nighttime low-peak periods is extracted to obtain a key road section-time period degradation feature set with spatiotemporal labels. Based on the key road section-time period degradation feature set, the coupling phenomenon of amplification or suppression of degradation contribution of different spatial road sections under different time periods is identified, and the joint modulation relationship of spatial features and time period features on degradation rate is obtained. Based on the joint modulation relationship, the line spatial degradation features and degradation time series curves are dynamically correlated and mapped to generate a composite correlation degree for quantifying the strength of brake disc degradation risk at any operating time and any line location. The calculation and matching logic of the composite correlation degree is used as the spatial-temporal correlation rule. The line spatial degradation features and degradation time series curves are bidirectionally coupled and positively correlated.

[0057] The critical road segment-time period degradation feature set can be a collection of key road segments with spatiotemporal labels, recording their degradation performance at different time periods. The joint modulation relationship can be the coupling relationship between the spatial road segment degradation contributions at different time periods. The composite correlation degree can be an indicator that quantifies the strength of brake disc degradation risk at any operating time and any line location. The bidirectional coupling positive correlation can be the relationship of mutual promotion and positive influence between the spatial degradation characteristics of the line and the degradation time series curve.

[0058] Specifically, in the operation of new energy buses, relying solely on single spatial or temporal degradation information analysis before constructing spatial-temporal association rules can lead to misjudgment of degradation risks. Degradation risks in high-load sections during peak hours are underestimated, easily causing braking failures and safety accidents. Over-maintenance of low-risk sections during off-peak hours results in resource waste and makes subsequent maintenance strategies less targeted. To address these issues, data fusion and feature engineering techniques are first employed to deeply integrate "brake disc spatial degradation information" (such as identified key sections like long downhill slopes and densely packed bus stops, and their spatial associations) with "brake disc temporal degradation information" (such as a degradation time-series curve with a baseline degradation rate of 0.8 during morning peak hours and 0.3 during off-peak hours). Through spatiotemporal cross-indexing, each key section is labeled with quantified degradation performance tags for different operating periods (morning peak, evening peak, etc.), thus constructing a structured "key section-time-period degradation feature set." Subsequently, pattern recognition and association analysis methods are applied to... This feature set is mined to identify the interaction patterns between spatial and temporal features. For example, cluster analysis reveals that "the degradation contribution of densely populated platform areas is significantly amplified during the morning rush hour, with its instantaneous risk value being several times higher than that of the same section during off-peak hours," while "during the nighttime off-peak hours, when passing through long downhill sections, the degradation contribution is significantly suppressed due to sparse traffic flow and gentle braking." This reveals the specific "joint modulation relationship." Finally, based on this relationship, a correlation function is designed using dynamic mapping and quantitative modeling to nonlinearly weight and fuse the "spatial degradation feature" curve with the "degradation time series curve." This function can dynamically calculate and output a quantitative "composite correlation degree" (e.g., 0.92) based on the real-time input vehicle GPS location (e.g., "between Xinhua Road Station and Cultural Palace Station") and timestamp (e.g., "8:15 AM"). This calculation and matching logic constitute the final "spatial-temporal correlation rule," enabling accurate and dynamic assessment of brake disc degradation risk at any operating time and any line location.

[0059] The method provided in this embodiment accurately captures the degradation pattern of the brake disc, avoids the bias of single-dimensional analysis, makes the maintenance guidance information more in line with actual operation, provides scientific and logical support for subsequent real-time load matching and maintenance behavior triggering, reduces ineffective maintenance, lowers operating costs, and ensures accurate maintenance timing.

[0060] In some embodiments, based on spatial-temporal association rules, real-time vehicle location data and time data are matched with the composite correlation degree corresponding to the current operating time and the location of the bus route to obtain a real-time composite correlation degree. Based on the real-time composite correlation degree, the comprehensive degradation state of the current brake disc caused by the combined effect of real-time operating load and historical accumulated degradation information is analyzed to obtain a real-time accumulated load spectrum. Based on the real-time accumulated load spectrum, maintenance warnings and execution instructions are triggered according to the current comprehensive degradation state, and maintenance levels, suggested maintenance time windows, and maintenance operation items for the corresponding bus are generated to obtain a vehicle maintenance information set.

[0061] Time data can be real-time time information during vehicle operation. Real-time composite correlation can be a quantitative indicator of the strength of brake disc degradation risk under the current operating state. Real-time operating load can be the instantaneous load intensity borne by the braking system at the current operating time and location on the route. Comprehensive degradation state can be the overall degradation degree and risk level of the brake disc under the combined effect of real-time operating load and historical accumulated degradation information. Real-time accumulated load spectrum can be a set of comprehensive degradation state assessment values ​​that integrate the impact of current real-time operating load and historical accumulated degradation results and are continuously updated with vehicle operation. Maintenance warning can be an alert message triggered based on the comprehensive degradation state, prompting relevant personnel to perform brake disc maintenance, and is divided into different urgency levels. Execution instructions can be operation instructions generated based on maintenance warnings to guide specific maintenance work. Maintenance level can be a maintenance urgency level divided according to the comprehensive degradation state of the brake disc. Recommended maintenance time window can be a time period recommended for maintenance work based on vehicle operation plan, route congestion, and maintenance resource allocation. Maintenance operation items can be specific maintenance actions that need to be performed for the current brake disc degradation state.

[0062] Specifically, during the operation of new energy buses, the risk of brake disc degradation dynamically changes with the route and time of day. Existing fixed-cycle maintenance is prone to "delayed maintenance" or "over-maintenance": vehicles that frequently travel down long slopes and experience high-intensity braking during peak hours experience rapid brake disc degradation but are difficult to maintain in a timely manner, easily leading to safety accidents; while for vehicles operating on smooth routes, over-maintenance increases operating costs and downtime. To address these issues: High-precision GPS location data and synchronized timestamps are continuously collected through onboard IoT terminals. For example, when a bus travels to a "long downhill section of Zhongshan Road with a 5% gradient and a length of 800 meters from north to south" at 5:30 PM (evening peak hour), this real-time data (location and time) is immediately uploaded to the cloud. The cloud maintenance engine then initiates a rule-matching and state fusion calculation process: First, it... Based on the real-time location and time, a quick query and mapping is performed in the pre-set "spatial-temporal association rule" knowledge base to calculate the "composite correlation degree" value (e.g., a high value, such as 0.85) corresponding to the specific combination of "downhill section" and "evening peak period". This is the "real-time composite correlation degree". Then, the vehicle's historical degradation file is called up, and a dynamic weight superposition algorithm is used to fuse the above-mentioned 0.85, which represents the instantaneous load intensity, with the historically accumulated degradation baseline value. This algorithm will consider the accelerating effect of the current high-intensity load on the existing "fatigue damage" and finally output a dynamically updated "real-time cumulative load spectrum" value that represents the overall wear and tear status from the start of service to the present. Finally, the decision logic compares this new value with the preset maintenance thresholds at various levels (e.g., the mild warning threshold of 0.6 and the moderate maintenance threshold of 0.8). Since the current value of 0.85 has exceeded the moderate maintenance threshold, a structured "vehicle maintenance information set" is automatically generated. The specific content includes: maintenance level "moderate maintenance", suggested execution time window "today's nighttime entry period", and operation item list "brake disc cleaning, thickness measurement and high temperature discoloration inspection". This information set is pushed to the fleet dispatch and maintenance work order system in real time to drive a precise predictive maintenance action.

[0063] The method provided in this embodiment avoids the safety hazards of untimely maintenance of critical vehicles, eliminates unnecessary over-maintenance, reduces operating costs and downtime losses, and generates personalized maintenance information to provide data support for subsequent strategy adjustments, promoting the shift of maintenance from "passive response" to "proactive prediction", and improving the level of maintenance intelligence and resource utilization.

[0064] In some embodiments, based on real-time composite correlation, the real-time operating load intensity level corresponding to the current operating time and the location of the bus route is analyzed to obtain the real-time load level; based on historical cumulative degradation information, the instantaneous load superposition effect and dynamic acceleration effect of the real-time load level on the historical cumulative degradation information are analyzed to obtain the real-time load action characteristics; based on the real-time load action characteristics, the historical cumulative degradation information is dynamically corrected to generate a comprehensive degradation state assessment value that integrates the impact of the current real-time operating load and the historical degradation accumulation results. The set of the comprehensive degradation state assessment value continuously updated with the operation of the bus and its impact on the brake disc is used as the real-time cumulative load spectrum.

[0065] Real-time load level can be a classification indicator of the operating load intensity of the brake disc corresponding to the current operating time and track location. Instantaneous load superposition effect can be the superposition effect of the current real-time load directly added to historical accumulated degradation information. Dynamic acceleration effect can be the accelerating effect of the current real-time load on the historical degradation rate. Comprehensive degradation status assessment value can be a quantitative indicator of the current actual degradation status of the brake disc.

[0066] Specifically, during the operation of new energy buses, if a real-time cumulative load spectrum is not constructed and only historical degradation data or instantaneous load data is relied upon, it will be impossible to accurately capture the dynamic correlation between real-time operating load and historical cumulative degradation. This will lead to misjudgment of the overall degradation status of the brake disc, resulting in excessive maintenance that increases operating costs, or delayed maintenance that ignores safety hazards. To address these issues: First, based on real-time matching "spatial-temporal correlation rules," the real-time composite correlation degree of the current moment and location (e.g., a value between 0 and 1) is output, and converted into a specific real-time load level through a preset mapping table (e.g., a correlation degree greater than 0.8 is mapped to "high load"). Next, the historical cumulative degradation information of the brake disc (e.g., a degradation index curve with time as the sequence) is obtained from the cloud database. The core processing step is to analyze the effect of the real-time load level on historical data: through a load impact model, the real-time load... The load level is converted into a "load modulation factor" and an "instantaneous load increment". For example, when the real-time load level is "high", the load modulation factor may be set to 1.5 (acceleration effect) and the instantaneous load increment may be set to 0.05 (superposition effect). Then, the historical degradation information is dynamically corrected: the degradation assessment value of the previous moment (e.g., 0.6) is added to the instantaneous load increment (0.05) to obtain the base value (0.65); at the same time, the degradation rate of the previous moment (e.g., 0.02 per hour) is multiplied by the load modulation factor (1.5) to obtain the modulated new degradation amount (0.03); finally, the base value (0.65) and the new degradation amount (0.03) are added to generate a brand-new comprehensive degradation state assessment value (0.68) for the current moment. This value is recorded in real time and added to the sequence. As the vehicle continues to run, this process is iterated in cycles, thereby forming a continuously dynamically updated real-time cumulative load spectrum.

[0067] The method provided in this embodiment accurately presents the actual degradation state of the brake disc, providing a reliable basis for maintenance early warning, level determination, and time window setting, effectively avoiding over-maintenance and under-maintenance problems, reducing operating costs, and providing data support for cluster maintenance scheduling, improving maintenance resource efficiency, and ensuring safe and stable vehicle operation.

[0068] In some embodiments, based on the vehicle maintenance information set, the clustering trend of maintenance needs among multiple vehicles and the timing of collaborative maintenance are analyzed according to the real-time cumulative load spectrum of different vehicles on different routes to obtain a cluster maintenance scheduling strategy. Based on the cluster maintenance scheduling strategy, the suggested maintenance time windows of each vehicle are dynamically optimized and merged to generate a collaborative maintenance execution sequence that takes into account both the urgency of individual vehicle degradation and the overall maintenance resource efficiency. According to the collaborative maintenance execution sequence, specific maintenance operation items are scheduled and executed, and the actual load characteristics, maintenance measures and effect verification data during the maintenance process are automatically recorded and integrated into the cloud database to form a structured predictive maintenance log for new energy vehicles.

[0069] The clustering trend of maintenance needs can be the concentrated characteristics of multiple vehicles in terms of maintenance time, maintenance level, and maintenance operation items. The timing of collaborative maintenance can be the optimal time to simultaneously meet the maintenance needs of multiple vehicles and maximize the overall utilization of maintenance resources. The cluster maintenance scheduling strategy can be a strategy plan for coordinating the maintenance work of multiple vehicles based on the clustering trend of multi-vehicle maintenance needs and the timing of collaborative maintenance. The urgency of single-vehicle degradation can be the urgency of the maintenance needs of a single vehicle due to brake disc degradation. Overall maintenance resource efficiency can be the utilization efficiency of human, material, and time resources during the maintenance process. The collaborative maintenance execution sequence can be the execution order formed after sorting the maintenance work of each vehicle under the guidance of the cluster maintenance scheduling strategy, clearly defining the maintenance priority, maintenance duration, and resource allocation scheme for each vehicle. Actual load characteristics can be the data characteristics related to the actual brake disc load detected during the maintenance process. Maintenance measures can be the actual maintenance operations and related auxiliary means performed. Effectiveness verification data can be data obtained after maintenance completion through vehicle trial operation testing, professional equipment testing, etc., used to verify the maintenance effect. A cloud database can be a remote cloud repository used to store various related data such as vehicle operation dynamics information, maintenance information, and effect verification data.

[0070] Specifically, in the operation of new energy bus clusters, the existing single-vehicle maintenance model ignores the correlation of brake disc degradation, resulting in dispersed maintenance needs and a situation where maintenance resources are both idle and strained. Furthermore, the dispersed downtime of vehicles affects route operation scheduling, with some vehicles experiencing accelerated degradation due to untimely maintenance and others undergoing premature maintenance, leading to resource waste. There is also a lack of full-process maintenance traceability data. To address these issues: First, spatiotemporal clustering analysis technology is used to perform multi-dimensional scanning of the "real-time cumulative load spectrum" and "suggested maintenance time window" of all vehicles. For example, density-based clustering algorithms are used to identify... By identifying multiple vehicles within similar time periods that are located at the same depot or on adjacent lines, a "maintenance demand cluster" is discovered, and a "cluster maintenance scheduling strategy" is generated accordingly. Subsequently, a dynamic window optimization algorithm is invoked to refine this strategy at the execution level. Its core is to use a merging strategy based on rules and heuristic search to intelligently adjust and merge the original time windows of each vehicle in the identified cluster. For example, three vehicles whose original maintenance windows are at 2:00 PM, 2:20 PM, and 2:45 PM respectively may be merged into a collaborative maintenance task block that starts at 2:00 PM and lasts for 90 minutes. During the merging process, each vehicle is weighted and sorted according to the degradation urgency quantified by its "real-time cumulative load spectrum," ensuring that vehicles with urgent needs receive a higher service priority in the merged task blocks. This ultimately generates a "collaborative maintenance execution sequence" that takes into account resource constraints (such as the number of workstations). This sequence is then automatically sent to the dispatch terminal in the repair shop, guiding the work teams to execute specific "maintenance operation items" in sequence. During each maintenance operation, maintenance personnel scan the vehicle code using a dedicated terminal, automatically linking it to the maintenance procedure and recording the operation. Simultaneously, by combining integrated sensors (such as torque wrenches and thermometers) with manual data entry, "actual load characteristics" (such as the actual thickness of the removed brake disc), "maintenance measures" (such as replacing brake pads of model XX), and "effect verification data" (such as the drag force test value after installation) are collected in real time. All these multimodal data are automatically encapsulated and uploaded. After data cleaning and association rule processing, they are structured and stored in a specific log table in the cloud database, forming a traceable and analyzable "predictive maintenance log for new energy vehicles."

[0071] The method provided in this embodiment optimizes the configuration of maintenance resources and avoids waste; merges maintenance time windows to reduce the total downtime of vehicles and ensure stable line capacity; structured logs enable full traceability of the maintenance process and provide data support for optimizing maintenance strategies; reduces the maintenance cost and waiting time per vehicle, extends the service life of brake discs, improves the safety and reliability of vehicle operation, and is suitable for large-scale operation scenarios.

[0072] Figure 3 This is a schematic diagram of the structure of a predictive maintenance system for new energy vehicles based on cloud data analysis, provided in one embodiment of this application. Figure 3As shown, the predictive maintenance system 300 for new energy vehicles based on cloud data analysis in this embodiment includes: a maintenance guidance analysis module 301, a cumulative load analysis module 302, and a maintenance control analysis module 303.

[0073] The maintenance guidance analysis module 301 is used to acquire a set of vehicle operation dynamic information, and based on the set of vehicle operation dynamic information, analyze the spatial-temporal correlation between brake disc degradation and bus route characteristics to obtain predictive maintenance guidance information; the cumulative load analysis module 302 is used to track vehicle operation behavior data based on the predictive maintenance guidance information, calculate the real-time cumulative load spectrum of the bus route, trigger personalized vehicle maintenance behavior, and obtain a set of vehicle maintenance information; the maintenance control analysis module 303 is used to dynamically adjust the bus maintenance control strategy based on the set of vehicle maintenance information and output the predictive maintenance log of new energy vehicles.

[0074] Optionally, when the maintenance guidance analysis module 301 analyzes the spatial-temporal correlation between brake disc degradation and bus route characteristics based on the vehicle operation dynamic information set to obtain predictive maintenance guidance information, it is specifically used for: the vehicle operation dynamic information set including historical vehicle braking data, route geographical information, and route operating timetable; based on the route geographical information, analyzing the differentiated impact of spatial characteristics of each segment of the bus operating route on the brake disc thermal load to obtain brake disc spatial degradation information; based on the vehicle historical braking data and combined with the route operating timetable, analyzing the temporal variation law of braking behavior of the vehicle in different operating segments of the same route to obtain brake disc temporal degradation information; integrating the brake disc spatial degradation information and the brake disc temporal degradation information to construct a spatial-temporal correlation rule reflecting the coupling between route characteristics and brake disc degradation to obtain the predictive maintenance guidance information.

[0075] Optionally, the maintenance guidance analysis module 301, during the construction of the brake disc spatial degradation information, is specifically used for: identifying and extracting long downhill sections, continuous curved sections, and densely packed platform sections in the line based on the line geographic information, as key sections for brake thermal load; analyzing the spatial synergistic effect and thermal accumulation effect of braking behavior between adjacent key sections based on the key sections for brake thermal load, to obtain brake load correlation information between sections; and analyzing the line spatial degradation characteristics reflecting the differences in the contribution of different spatial locations to brake disc thermal load based on the key sections for brake thermal load and the brake load correlation information between sections, to obtain the brake disc spatial degradation information.

[0076] Optionally, the maintenance guidance analysis module 301, during the construction of the brake disc time-series degradation information, is specifically used for: based on the route operating timetable and combined with the vehicle's historical braking data, analyzing the distribution characteristics of bus braking frequency and braking intensity during each period of morning peak, evening peak, off-peak, and nighttime low-peak to obtain time-series braking information; based on the time-series braking information, analyzing the differentiated effects of frequent starts and stops and high-intensity braking during peak hours, medium-intensity braking during stable operation during off-peak hours, and low-intensity braking during low-peak hours on brake disc thermal fatigue accumulation to obtain time-series braking load characteristics; based on the time-series braking information and the time-series braking load characteristics, analyzing the degradation time-series curve reflecting the change of brake disc degradation rate with time bands to obtain the brake disc time-series degradation information.

[0077] Optionally, when analyzing the degradation time-series curve reflecting the change of brake disc degradation rate with time bands, the maintenance guidance analysis module 301 is specifically used to: based on the time-series braking load characteristics, analyze the baseline level of brake disc degradation rate in each period of the morning peak, the evening peak, the off-peak period, and the nighttime low-peak period, as well as the step change law of the baseline level as the operating period switches, to obtain the degradation rate time-series baseline spectrum; based on the degradation rate time-series baseline spectrum, combined with the time-series braking information, analyze the inheritance effect of the accumulated thermal fatigue of the previous operating period on the initial degradation state of the next period, as well as the dynamic modulation effect of the inheritance effect on the degradation rate of the next period, to obtain the inter-period degradation rate modulation relationship; and construct the complete degradation time-series curve reflecting the continuous and adaptive change of brake disc degradation rate in different time bands based on the degradation rate time-series baseline spectrum and the inter-period degradation rate modulation relationship.

[0078] Optionally, the maintenance guidance analysis module 301, during the construction of the spatial-temporal association rule, is specifically used for: based on the brake disc spatial degradation information and combined with the brake disc temporal degradation information, extracting the degradation performance of the key road segments of the brake heat load during the morning peak, the evening peak, the off-peak period, and the nighttime low peak period, to obtain a key road segment-time period degradation feature set with spatiotemporal labels; based on the key road segment-time period degradation feature set, identifying the coupling phenomenon where the degradation contribution of different spatial road segments is amplified or suppressed under different time periods, to obtain the joint modulation relationship between spatial features and time period features on the degradation rate; based on the joint modulation relationship, dynamically associating and mapping the line spatial degradation features with the degradation time series curve to generate a composite correlation degree used to quantify the strength of brake disc degradation risk at any operating time and any line position, using the calculation and matching logic of the composite correlation degree as the spatial-temporal association rule; the line spatial degradation features and the degradation time series curve are bidirectionally coupled and positively correlated.

[0079] Optionally, during the construction of the vehicle maintenance information set, the cumulative load analysis module 302 is specifically used to: according to the spatial-temporal association rules, real-time vehicle location data and time data are obtained, and the composite correlation degree corresponding to the current running time and the location of the bus route is matched to obtain the real-time composite correlation degree; based on the real-time composite correlation degree, the comprehensive degradation state of the current brake disc formed by the combined effect of real-time operating load and historical cumulative degradation information is analyzed to obtain the real-time cumulative load spectrum; based on the real-time cumulative load spectrum, maintenance warnings and execution instructions are triggered according to the current comprehensive degradation state, and maintenance levels, suggested maintenance time windows and maintenance operation items are generated for the corresponding bus vehicles to obtain the vehicle maintenance information set.

[0080] Optionally, the cumulative load analysis module 302, during the construction of the real-time cumulative load spectrum, is specifically used for: analyzing the real-time operating load intensity level corresponding to the current operating time and the location of the bus route based on the real-time composite correlation degree, to obtain the real-time load level; analyzing the instantaneous load superposition effect and dynamic acceleration effect of the real-time load level on the historical cumulative degradation information based on the historical cumulative degradation information, to obtain the real-time load effect characteristics; dynamically correcting the historical cumulative degradation information based on the real-time load effect characteristics, generating a comprehensive degradation state evaluation value that integrates the impact of the current real-time operating load and the historical degradation accumulation results, and using the set of the comprehensive degradation state evaluation value continuously updated with the operation of the bus to affect the brake disc as the real-time cumulative load spectrum.

[0081] Optionally, when the maintenance control analysis module 303 dynamically adjusts the bus maintenance control strategy based on the vehicle maintenance information set and outputs predictive maintenance logs for new energy vehicles, it is specifically used for: based on the vehicle maintenance information set, analyzing the aggregation trend of maintenance needs and collaborative maintenance timing among multiple vehicles according to the real-time cumulative load spectrum of different vehicles on different routes, and obtaining a cluster maintenance scheduling strategy; based on the cluster maintenance scheduling strategy, dynamically optimizing and merging the suggested maintenance time windows of each vehicle to generate a collaborative maintenance execution sequence that takes into account both the urgency of individual vehicle degradation and the overall maintenance resource efficiency; scheduling and executing specific maintenance operation items according to the collaborative maintenance execution sequence, and automatically recording the actual load characteristics, maintenance measures, and effect verification data during the maintenance process, integrating them into the cloud database to form a structured predictive maintenance log for new energy vehicles.

[0082] The system in this embodiment can be used to execute the methods of any of the above embodiments, and its implementation principle and technical effect are similar, so they will not be described again here.

Claims

1. A predictive maintenance method for new energy vehicles based on cloud data analysis, characterized in that, include: Obtain a set of vehicle operation dynamic information; based on the set of vehicle operation dynamic information, analyze the spatial-temporal correlation between brake disc degradation and bus route characteristics to obtain predictive maintenance guidance information. The vehicle operation dynamic information set uses onboard sensors mounted on new energy buses as the data source; Based on the predictive maintenance guidance information, vehicle operation behavior data is tracked, the real-time cumulative load spectrum of the bus route is calculated, personalized vehicle maintenance behavior is triggered, and a vehicle maintenance information set is obtained. Based on the vehicle maintenance information set, the bus maintenance control strategy is dynamically adjusted, and predictive maintenance logs for new energy vehicles are output. Based on the vehicle operation dynamic information set, the spatial-temporal correlation between brake disc degradation and bus route characteristics is analyzed to obtain predictive maintenance guidance information, including: The vehicle operation dynamic information set includes historical vehicle braking data, route geographic information, and route operation timetable. Based on the aforementioned route geographic information, the spatial characteristics of each section of the bus operation route are analyzed to determine the differential impact on the brake disc thermal load, thereby obtaining brake disc spatial degradation information. Based on the vehicle's historical braking data and the route's operating timetable, the temporal variation pattern of the vehicle's braking behavior during different operating periods on the same route is analyzed to obtain brake disc temporal degradation information. By integrating the brake disc spatial degradation information and the brake disc temporal degradation information, a spatial-temporal association rule reflecting the coupling between line characteristics and brake disc degradation is constructed to obtain the predictive maintenance guidance information; The process of constructing the spatial-temporal association rules includes: Based on the spatial degradation information of the brake disc and combined with the temporal degradation information of the brake disc, the degradation performance of key road sections under braking thermal load during morning peak, evening peak, off-peak and nighttime low-peak periods is extracted to obtain a key road section-time period degradation feature set with spatiotemporal labels. Based on the aforementioned key road segment-time period degradation feature set, the coupling phenomenon in which the degradation contribution of different spatial road segments is amplified or suppressed under different time periods is identified, and the joint modulation relationship between spatial features and time period features on degradation rate is obtained. Based on the joint modulation relationship, the spatial degradation characteristics of the line are dynamically correlated and mapped with the degradation time-series curve to generate a composite correlation degree for quantifying the strength of brake disc degradation risk at any operating time and any line position. The calculation and matching logic of the composite correlation degree is used as the spatial-temporal correlation rule. The spatial degradation characteristics of the line are bidirectionally coupled and positively correlated with the degradation time-series curve; The process of constructing the vehicle maintenance information set includes: Based on the spatial-temporal association rules, the real-time vehicle location data and time data are matched with the composite association degree corresponding to the current running time and the location of the bus route to obtain the real-time composite association degree. Based on the real-time composite correlation, the comprehensive degradation state of the current brake disc caused by the combined effect of real-time operating load and historical cumulative degradation information is analyzed to obtain the real-time cumulative load spectrum. Based on the real-time cumulative load spectrum, maintenance warnings and execution instructions are triggered according to the current comprehensive degradation status, and maintenance levels, suggested maintenance time windows, and maintenance operation items are generated for the corresponding buses to obtain the vehicle maintenance information set; The process of constructing the real-time cumulative load spectrum includes: Based on the real-time composite correlation, the real-time operating load intensity level corresponding to the current operating time and the location of the bus route is analyzed to obtain the real-time load level; Based on the historical cumulative degradation information, the instantaneous load superposition effect and dynamic acceleration effect of the real-time load level on the historical cumulative degradation information are analyzed to obtain the real-time load effect characteristics. Based on the real-time load characteristics, the historical cumulative degradation information is dynamically corrected to generate a comprehensive degradation status assessment value that integrates the impact of the current real-time operating load and the historical degradation accumulation results. The set of the comprehensive degradation status assessment value that continuously updates the impact on the brake disc as the bus runs is used as the real-time cumulative load spectrum.

2. The method according to claim 1, characterized in that, The process of constructing the brake disc space degradation information includes: Based on the geographical information of the route, long downhill sections, continuous curve sections, and densely packed platform sections in the route are identified and extracted as key sections for braking heat load. Based on the key road segments with braking thermal load, the spatial synergistic effect and thermal accumulation effect of braking behavior between adjacent key road segments are analyzed to obtain braking load correlation information between road segments. Based on the key road sections of the braking thermal load and the braking load correlation information between the road sections, the spatial degradation characteristics of the line, which reflect the differences in the contribution of different spatial locations to the braking disc thermal load, are analyzed to obtain the spatial degradation information of the braking disc.

3. The method according to claim 2, characterized in that, The process of constructing the brake disc timing degradation information includes: Based on the route timetable and combined with the vehicle's historical braking data, the distribution characteristics of bus braking frequency and braking intensity during the morning peak, evening peak, off-peak, and nighttime low-peak periods are analyzed to obtain time-specific braking information. Based on the time-series braking information, the differential effects of frequent starts and stops and high-intensity braking during peak hours, medium-intensity braking during stable operation during off-peak hours, and low-intensity braking during low-peak hours on the thermal fatigue accumulation of the brake disc are analyzed to obtain the time-series braking load characteristics. Based on the time-dependent braking information and the time-sequential braking load characteristics, the degradation time-sequential curve reflecting the change of the brake disc degradation rate with time band is analyzed to obtain the time-sequential degradation information of the brake disc.

4. The method according to claim 3, characterized in that, The analysis reflects the degradation time series curves that reflect the change in brake disc degradation rate over time, including: Based on the time-series braking load characteristics, the baseline level of brake disc degradation rate during each period of the morning peak, the evening peak, the off-peak period and the nighttime low peak period is analyzed, as well as the step change law of the baseline level as the operating period switches, to obtain the time-series baseline spectrum of degradation rate. Based on the degradation rate time period reference spectrum and combined with the time period braking information, the inheritance effect of the thermal fatigue accumulated in the previous running period on the initial degradation state of the next period, and the dynamic modulation effect of the inheritance effect on the degradation rate of the next period are analyzed to obtain the degradation rate modulation relationship between time periods. Based on the degradation rate time-period reference spectrum and the degradation rate modulation relationship between time periods, a degradation time-series curve that fully reflects the continuous and adaptive changes in the brake disc degradation rate in different time bands is constructed.

5. The method according to claim 4, characterized in that, The process of dynamically adjusting the bus maintenance control strategy based on the vehicle maintenance information set and outputting predictive maintenance logs for new energy vehicles includes: Based on the vehicle maintenance information set, and according to the real-time cumulative load spectrum of different vehicles on different routes, the aggregation trend of maintenance needs among multiple vehicles and the timing of collaborative maintenance are analyzed to obtain a cluster maintenance scheduling strategy. Based on the cluster maintenance scheduling strategy, the suggested maintenance time windows for each vehicle are dynamically optimized and merged to generate a collaborative maintenance execution sequence that takes into account both the urgency of individual vehicle degradation and the overall maintenance resource efficiency. According to the collaborative maintenance execution sequence, specific maintenance operation items are scheduled and executed, and the actual load characteristics, maintenance measures and effect verification data during the maintenance process are automatically recorded and integrated into the cloud database to form a structured predictive maintenance log for the new energy vehicle.

6. A predictive maintenance system for new energy vehicles based on cloud data analysis, characterized in that: The method applied to any one of claims 1-5 includes: The maintenance guidance analysis module is used to acquire a set of vehicle operation dynamic information, and based on the set of vehicle operation dynamic information, analyze the spatial-temporal correlation between brake disc degradation and bus route characteristics to obtain predictive maintenance guidance information. The cumulative load analysis module is used to track vehicle operation behavior data based on the predictive maintenance guidance information, calculate the real-time cumulative load spectrum of the bus route, trigger personalized vehicle maintenance behavior, and obtain a vehicle maintenance information set. The maintenance control analysis module is used to dynamically adjust the bus maintenance control strategy based on the vehicle maintenance information set and output the predictive maintenance log of new energy vehicles.

Images