Real-time dynamic power distribution optimization method for electric vehicle controller

Abstract

The invention relates to the technical field of electric vehicle power control, and provides a real-time dynamic power distribution optimization method for an electric vehicle controller. The method comprises the steps that after an input destination is received, a navigation route is generated according to the real-time position of the electric tricycle and the destination; performing global road condition analysis on the navigation route to obtain M local road sections and road condition information; the interactive load sensor obtains a real-time load; analyzing a power demand according to the real-time load and the M pieces of local road condition information, and outputting M power distribution modes; and when the electric vehicle controller runs along the navigation route, power distribution dynamic optimization and mode smooth switching are executed according to a local road section connection sequence until the electric vehicle arrives at a destination. The technical problems of low power distribution efficiency and insufficient cruising ability of the electric vehicle under complex road conditions and dynamic loads are solved, and the effects of automatically adapting to different road sections and load conditions through global road condition analysis and load sensing and improving the power distribution efficiency and the cruising ability of the electric vehicle are achieved.

Description

Technical Field

[0001] The present application relates to the technical field of electric vehicle power control, and in particular to a real-time dynamic power allocation optimization method for an electric vehicle controller. Background Art

[0002] There are many deficiencies in the power distribution of existing electric vehicle control systems under complex road conditions and dynamic load conditions. Traditional electric vehicle control systems usually rely on fixed power distribution strategies. This single control method cannot adjust the power output in real time according to changes in actual driving conditions, resulting in low energy efficiency utilization, insufficient endurance and poor driving experience of electric vehicles under variable road conditions or load conditions. Especially in the face of complex urban traffic environments, road conditions are often complex and changeable, such as slope changes, traffic flow fluctuations and differences in road friction coefficients, making it difficult for fixed power distribution strategies to adapt, and it is easy to have insufficient power or power waste. In addition, in the existing technology, the load sensing capability is limited, and it is difficult to accurately capture the real-time load status of the electric tricycle, which further limits the implementation of power distribution optimization. Even if some technologies try to adapt to a variety of scenarios by setting different power modes, the switching of these modes often relies on manual settings and lacks dynamic optimization capabilities. The mode switching process may be uneven, affecting the ride comfort and even increasing the unit energy consumption of the electric tricycle. Summary of the invention

[0003] This application provides a real-time dynamic power allocation optimization method for an electric vehicle controller, aiming to solve the technical problems of low power allocation efficiency and insufficient endurance of electric vehicles under complex road conditions and dynamic loads.

[0004] In view of the above problems, the present application provides a real-time dynamic power allocation optimization method for an electric vehicle controller.

[0005] The present application provides a real-time dynamic power distribution optimization method for an electric vehicle controller, the method comprising: after receiving a destination input by a target user, generating a navigation route according to the real-time position of an electric tricycle and the destination; performing a global road condition analysis on the navigation route to obtain M local sections and M local road condition information, wherein the M local sections are connected to restore the navigation route; an interactive load sensor obtains the real-time load of the electric tricycle; performing a power distribution demand analysis according to the real-time load and the M local road condition information, and outputting M power distribution modes; the electric vehicle controller, during the electric tricycle traveling along the navigation route, correspondingly performs intra-mode dynamic optimization of power distribution and smooth switching between modes of the M power distribution modes according to the connection order of the M local sections until the electric tricycle arrives at the destination.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: For the above real-time dynamic power distribution optimization method of an electric vehicle controller, the method generates a complete navigation route by receiving the destination input by the user and combining the real-time position of the electric tricycle, ensuring the accuracy of the driving path planning; subsequently, it analyzes the global road conditions of the generated navigation route, refines it into M local road sections, and extracts the corresponding road condition information for each local road section, thereby realizing the refined decomposition of complex road conditions and providing accurate data support for subsequent power distribution. At the same time, it interacts with the load sensor to sense the load change of the electric tricycle in real time, ensuring the dynamic update and accurate capture of the load data; in terms of power distribution, combining the real-time load and local road condition information, M power distribution modes are generated through demand analysis to accurately match the driving requirements of each road section. During the driving process of the electric tricycle, the electric vehicle controller strictly executes the corresponding power distribution modes one by one according to the connection sequence of the local road sections. The implementation of dynamic optimization within the mode and smooth switching between modes not only improves the accuracy of power distribution but also effectively avoids the power interruption or fluctuation phenomenon caused by uneven mode switching in the traditional method, ensuring the smoothness of the electric tricycle driving and driving comfort.

[0007] The above description is only an overview of the technical solution of this application. In order to be able to understand the technical means of this application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of this application more obvious and understandable, the specific embodiments of this application are specifically given below. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0009] Figure 1 It is a schematic flowchart of a real-time dynamic power distribution optimization method of an electric vehicle controller in an embodiment.

[0010] Figure 2 It is a schematic flowchart of the dynamic optimization of power distribution and smooth switching between modes of a real-time dynamic power distribution optimization method of an electric vehicle controller in an embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0011] By providing a real-time dynamic power distribution optimization method for an electric vehicle controller in the embodiments of this application, the technical problems of low power distribution efficiency and insufficient battery life of electric vehicles under complex road conditions and dynamic loads are solved.

[0012] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0013] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products, or devices.

[0014] Embodiment, such as Figure 1 As shown, the present application provides a real-time dynamic power distribution optimization method for an electric vehicle controller, and the method includes: After receiving the destination input by the target user, generate a navigation route according to the real-time position of the electric tricycle and the destination.

[0015] In the embodiments of the present application, first, receive the destination information input by the user through the interaction interface, and then obtain the real-time position information of the electric tricycle. These information are usually provided by the GPS module to ensure the accuracy of the position data; based on the real-time position of the electric tricycle and the destination specified by the user, call the map service interface (such as map API) for path planning to generate an optimal navigation route. The generated navigation route not only contains the global path information, but also marks the key nodes along the way and the driving characteristics that need attention, providing basic data support for subsequent power distribution optimization and dynamic adjustment. Through this process, ensure that the electric tricycle can reach the target position according to the optimal path, while reducing energy consumption and driving time.

[0016] By performing a global road condition analysis on the navigation route, obtain M local road sections and M local road condition information, where the M local road sections are connected to restore the navigation route.

[0017] In one embodiment, after generating a navigation route, a global road condition analysis is performed on the entire route. This process obtains the overall road condition information of the navigation route by calling a map service interface (such as a map API), including detailed data such as traffic flow, road slope, elevation change, and road surface conditions. Subsequently, in combination with the road condition characteristics, the entire navigation route is split into fine-grained segments, divided into M local road segments, and each local road segment corresponds to a specific set of road condition information, including the traffic condition, road slope, elevation change, and road surface conditions of this segment. This segmented analysis can not only clearly restore the overall structure of the navigation route but also capture the specific factors that may affect the driving performance of the electric tricycle in each section of the journey. Through the connection of the M local road segments and their road condition information, the restored navigation route maintains integrity in structure and at the same time provides an accurate support basis for subsequent power distribution. In this way, it is possible to more precisely optimize the power distribution according to the actual conditions of different road segments, thereby improving the energy efficiency utilization rate and driving stability of the electric tricycle.

[0018] Further, the present application provides a method for obtaining M local road segments and M local road condition information through global road condition analysis of the navigation route, and the method includes: Obtaining the global road condition information of the navigation route by calling the map API; traversing the global road condition information by using the power distribution correlation characteristic set, and locating multiple sets of correlation characteristic change nodes on the navigation route; finely splitting the navigation route according to the multiple sets of correlation characteristic change nodes to obtain the M local road segments; and decomposing the global road condition information with the M local road segments as constraints to obtain M local road condition information.

[0019] Preferably, when performing global road condition analysis, first, by calling a map API (such as Baidu Map, Amap, or Google Map API), the global road condition information of the navigation route is obtained. This information includes the real-time traffic flow of the road (such as the degree of congestion), road slope, elevation change (such as altitude change), road surface condition, road type (such as highway, main road, or branch road), etc. This process provides comprehensive data support for subsequent road condition analysis. Subsequently, according to the predefined power distribution correlation feature set (such as traffic flow feature, slope feature, elevation feature, and friction feature), the global road condition information of the navigation route is traversed segment by segment to detect the feature change points. For example, when the slope changes significantly or the traffic flow changes from unobstructed to congested, these change points are marked as associated feature change nodes, which can indicate the key positions where the power demand of the electric tricycle may change significantly. After that, with the marked associated feature change nodes as boundaries, the navigation route is split in a fine-grained manner (such as being divided according to the change of traffic flow, divided according to the slope of the road, etc.). This fine-grainedness indicates the type of change that can be used for route splitting. Each segment after splitting is called a local segment, and the range of each local segment is determined by two associated feature change nodes that meet the fine-grained requirements before and after. This method ensures that the road condition information of each local segment has high consistency, facilitating the precise optimization of subsequent power distribution. For each local segment, the corresponding road condition information is extracted from the global road condition information to form a local road condition information set that corresponds one-to-one with the local segment. These local road condition information include detailed data such as traffic flow, road slope, and elevation change within the segment. Each local segment and its corresponding local road condition information together constitute a refined representation of the entire navigation route, providing accurate basic data support for the subsequent generation and dynamic optimization of the power distribution mode.

[0020] The interactive load sensor obtains the real-time load of the electric tricycle; based on the real-time load and M local road condition information, power distribution demand analysis is performed, and M power distribution modes are output.

[0021] In one embodiment, by interacting with the load sensor, the load information of the electric tricycle is collected in real time, including the current load weight and dynamic changes (such as load fluctuations caused by factors such as acceleration, deceleration, or tilting). These real-time load data can accurately reflect the actual load state of the electric tricycle during operation, providing a basic basis for power distribution. Based on the obtained real-time load data, combined with M local road condition information (such as traffic flow, slope size, etc.) parsed from the navigation route, the power demand for each local section is analyzed, that is, M power distribution modes are matched from the mode database according to the characteristics of each section of road condition and real-time load. Each power distribution mode includes an output power range, a torque range, a current range, a power distribution ratio range, etc. It should be noted that since the road condition information of different local sections may be similar, for example, the slopes and traffic flows of some sections are similar, so these sections may match the same power distribution mode. That is to say, there may be duplicate modes among the generated M power distribution modes. These power distribution modes provide precise power adjustment support for the electric tricycle, ensuring the flexibility and rationality of power distribution during driving.

[0022] Further, the present application provides an interaction with a load sensor to obtain the real-time load of the electric tricycle, and analyzes the power distribution requirements according to the real-time load and M local road condition information, and outputs M power distribution modes, including: Interactively obtain a sample load set, a sample road condition information set, and a sample dynamic load set; perform multiple regression analysis on the sample load set, the sample road condition information set, and the sample dynamic load set, and construct a dynamic load function based on the analysis results; load the real-time load and M local road condition information into the dynamic load function for dynamic load calculation and update to obtain M local dynamic loads; pre-construct a mode database; load the M local dynamic loads and M local road condition information into the mode database for real-time mode matching, and output the M power distribution modes.

[0023] Preferably, through historical data collection and interaction, a sample load set, a sample road condition information set, and a sample dynamic load set including various working conditions are obtained. These data reflect the operating characteristics of the electric tricycle under different load conditions (such as light load, heavy load), different road condition characteristics (such as flat sections, slopes, slippery roads), and dynamic loads (such as acceleration, braking, or road surface friction changes), providing basic data for subsequent function construction. Subsequently, taking the above sample load set and sample road condition information set as independent variables and the sample dynamic load set as the target variable for multiple regression analysis, an initial dynamic load function is established through a multiple regression model. The form of this multiple regression model is ; where Y is the dynamic load value, is the intercept, , ,..., is the regression coefficient, representing the influence of each variable on the dynamic load value. , ,..., are the input independent variables, such as load, road condition information, etc. is the error term; after establishing the initial dynamic load function, the sample load set, sample road condition information set, and sample dynamic load set are divided into a training set and a test set (the division ratio is 8:2). Then, the divided training set is input into the multiple regression model, the model is fitted through the training set, and methods such as the ordinary least squares method (OLS) are used to optimize the regression coefficient and reduce the error term. After the training is completed, the test set is used to evaluate the model performance and calculate relevant indicators, such as the coefficient of determination, to quantify the explanatory ability of the model for the dynamic load. If the quantification result does not meet the expected expectation, the model is optimized. For example, the significance of the regression coefficient is analyzed, and variables with less or insignificant influence on the dynamic load value are removed (such as using a p-value test to remove variables with a p-value greater than 0.05). If it meets the expected expectation, the current dynamic load function is output; after constructing the dynamic load function, the collected real-time load and the parsed M local road condition information are loaded into the dynamic load function for dynamic load calculation, so as to obtain M local dynamic load values. These dynamic load values provide accurate reference data for the power distribution of each local road section; then, the calculated M local dynamic load values are loaded into a pre-constructed pattern database, which contains K power distribution patterns, such as an economy mode, a load mode, a sports mode, etc. Each mode corresponds to a specific output power range, torque range, current range, power distribution ratio, and power priority setting; after loading the M local dynamic load values and the M local road condition information into the pattern database, real-time pattern matching is performed through a decision tree connected to the pattern database. The decision tree judges the classification node to which the input data belongs layer by layer according to the dynamic load and road condition characteristics, combined with the dynamic load range and road condition characteristic range, and finally locates the best-matched power distribution pattern. For example, if the dynamic load value of a local road section is high and the slope is steep, the decision tree will match the load mode; if the dynamic load value is low and the road condition is flat, the economy mode will be matched; through this decision tree based on data training, the relationship between the input features and the power distribution pattern can be efficiently identified, and the optimal power distribution pattern can be matched in real time for each local road section, realizing the optimization of the electric vehicle energy efficiency and the improvement of the driving experience.

[0024] Furthermore, the present application provides a pre-constructed pattern database, and the method includes: Solidify K power distribution modes, where each power distribution mode includes an output power range, a torque range, a current range, a power distribution ratio range, and a power priority setting; call the historical power log of the electric tricycle, where the historical power log records multiple historical dynamic loads, multiple historical road conditions, multiple historical power distribution information, and multiple historical unit power consumptions, and the historical power distribution information includes output power records, torque records, current records, and power distribution ratio records; according to the interval matching relationship between the multiple historical power distribution information and the K power distribution modes, group the multiple historical dynamic loads, multiple historical road conditions, and multiple historical power distribution information to obtain K groups of associated historical dynamic loads, K groups of historical road conditions, and K groups of historical unit power consumptions; solve the power probability density of the K groups of historical unit power consumptions to obtain K power consumption fluctuation ranges; based on the K power consumption fluctuation ranges, perform mode fitness analysis on the K groups of historical dynamic loads and K groups of historical road conditions to obtain K dynamic load ranges and K groups of road condition characteristic ranges; 5 Based on the knowledge graph, associate and store the K dynamic load ranges, K groups of road condition characteristic ranges, and K power distribution modes to complete the construction of the mode database.

[0025] Optionally, according to the configuration of the electric tricycle, K kinds of power distribution modes are solidified, each mode corresponds to a specific output power range, torque range, current range, power distribution ratio range and power priority setting. For example, the output power range of the economic mode is 0.5-1.5kW, the torque range is 10-30Nm, the maximum speed is 30km / h, and the power priority is energy priority. It is suitable for scenarios where the electric tricycle has a light load and good road conditions. The goal is to complete the driving with the lowest energy consumption and maximize energy efficiency. The output power range of the load-carrying mode is 1.5-3.0kW, the torque range is 30-70Nm, the maximum speed is 25km / h, and the power priority is energy priority. The power priority is torque priority, which is suitable for electric tricycles with heavy loads or complex road conditions. The goal is to provide stronger torque support to ensure power output when loaded. The output power range of the sports mode is 2.0-5.0kW, the torque range is 50-100Nm, and the maximum speed is 50km / h. The power priority is power priority, which is suitable for electric tricycles with light loads and high speeds. The goal is to improve the power response and speed of electric tricycles. The special mode dynamically adjusts the output power, torque and speed. The power priority is continuous priority, which is suitable for long-term continuous climbing, large load fluctuations or complex working conditions (such as slope>8%, friction coefficient<0.6) For special scenarios, the goal is to dynamically adjust according to real-time working conditions to ensure continuous and stable output. Subsequently, power logs are extracted from the historical operation data of the electric tricycle. The logs record multiple historical dynamic loads, multiple historical road conditions, multiple historical power distribution information, and multiple historical unit power consumptions. Among them, the historical power distribution information includes output power records, torque records, current records, and power distribution ratio records. These data cover the operation performance of the electric tricycle under different working conditions and provide basic support for subsequent analysis. Then, according to the output power intervals, torque intervals, current intervals, etc. of the K power distribution modes, and in accordance with the interval matching rules, multiple historical dynamic loads, multiple historical road conditions, and multiple historical power distribution information are classified. Each group of data corresponds to a power distribution mode, forming K groups of historical dynamic loads, K groups of historical road conditions, and K groups of historical unit power consumptions. After that, statistical analysis is performed on the K groups of historical unit power consumption data. The K groups of historical unit power consumption data are divided into several intervals. The number of intervals can be obtained by taking the square root of the total number of data and rounding up. Then, the frequency density of each interval is calculated by using the ratio of the number of data points in each interval to the product of the interval width and the total number of data. A histogram is drawn based on the frequency density of each interval, and the histogram is normalized so that the area is 1, thereby obtaining K probability density functions. Next, the power consumption fluctuation intervals of each group of power distribution modes are calculated according to the K probability density functions. Usually, a 95% confidence interval is selected as the fluctuation range. The cumulative probability is calculated through the cumulative distribution curves (CDF, Cumulative Distribution Function) of the K probability density functions, and the corresponding upper and lower bounds are extracted from the cumulative distribution curves of the K probability density functions with the set confidence interval, obtaining K power consumption fluctuation intervals. For example, the data range between the cumulative probability from 0.025 to 0.975 is the power consumption fluctuation interval. Then, according to the K power consumption fluctuation intervals, the adaptability of each group of historical dynamic loads and historical road conditions is analyzed, and K dynamic load intervals and K groups of road condition characteristics intervals are extracted through multiple extreme value calls. Finally, the dynamic load intervals and road condition characteristics are used as the associated attributes of the power distribution mode. By establishing the mapping relationship between the dynamic load values, road condition characteristics (such as slope, friction coefficient) and the mode, a complete knowledge graph structure is formed. Through this associated storage, the knowledge graph can quickly retrieve and match the relevant conditions of the power mode and provide efficient support during real-time operation. Through the above process, the construction of the power distribution mode database based on historical operation data is completed, providing efficient and reliable support for the real-time power distribution optimization of the electric tricycle.

[0026] Furthermore, the present application provides a method for analyzing the mode fitness of the K groups of historical dynamic loads and K groups of historical road conditions according to the K power consumption fluctuation intervals, obtaining K dynamic load intervals and K groups of road condition characteristics intervals. The method includes: Based on the power consumption deviation between the K power consumption fluctuation intervals and the K sets of unit power consumption, perform reverse screening on the K sets of historical dynamic loads and K sets of historical road conditions information to obtain K sets of screened dynamic loads and K sets of screened road conditions information; perform load extreme value calls on the K sets of screened dynamic loads to obtain the K dynamic load intervals; pre-define a power distribution association characteristic set, where the power distribution association characteristic set includes traffic flow characteristics, slope characteristics, elevation characteristics, and friction characteristics; use the power distribution association characteristic set to perform multi-index extreme value calls on the K sets of screened road conditions information to obtain the K sets of road condition characteristic intervals.

[0027] Optionally, according to the calculated K power consumption fluctuation intervals, calculate the power consumption deviation from each set of unit power consumption data to evaluate the adaptability of each set of data in the power distribution mode. By calculating the deviation between each set of unit power consumption data and the fluctuation interval, screen out the historical dynamic loads and road condition information that conform to the power consumption fluctuation characteristics. The screening method includes excluding data with deviations exceeding a set threshold (such as 5% of the upper and lower limits), and retaining the adaptable dynamic loads and road condition information, thereby forming K sets of screened dynamic loads and K sets of screened road condition information. For example, if the power consumption fluctuation interval of the load mode is 1.5 - 2.5 kW / kg, then exclude data points with power consumption values less than 1.425 or greater than 2.625; for the K sets of dynamic load data screened out, perform load extreme value calls, that is, extract the maximum and minimum values in each set of dynamic loads, and determine the dynamic load interval based on the extracted maximum and minimum values; in addition, pre-define a power distribution association characteristic set, which includes traffic flow characteristics, slope characteristics, elevation characteristics, and friction characteristics. When screening the road condition information, analyze the screened road condition data based on this characteristic set, and extract the key road condition characteristics through the multi-index extreme value call method. For example, for the load mode, call the slope characteristic and determine the steep slope condition (slope > 5%), and at the same time select the road condition data with a friction coefficient less than 0.7 in combination with the friction characteristic, and finally obtain the road condition characteristic interval of the load mode; through the above process, generate K dynamic load intervals and K sets of road condition characteristic intervals, providing accurate support data for the subsequent construction of the knowledge graph and real-time power distribution.

[0028] During the process of the electric tricycle driving along the navigation route, the electric vehicle controller correspondingly executes the in-mode power distribution dynamic optimization and inter-mode smooth switching of the M power distribution modes according to the connection sequence of the M local road sections until the electric tricycle reaches the destination.

[0029] In one embodiment, during the process of the electric tricycle traveling along the navigation route, the electric vehicle controller will execute the power distribution mode matched for each local section according to the connection sequence of each local section in the navigation route. For each section, the controller will perform dynamic optimization according to the real-time state of the electric tricycle (such as battery state, driving state, etc.) under the current power distribution mode, and adjust power parameters such as output power and torque to adapt to the driving conditions of this section; when the electric tricycle enters the next local section from one local section, it will smoothly switch the power distribution mode on the premise of maintaining smooth driving, avoiding sudden changes in the power output of the electric tricycle due to mode switching. For example, when the electric tricycle enters a steep slope section from a flat section, the controller will smoothly switch from the economy mode to the load mode, and at the same time dynamically adjust the torque and output power to meet the climbing requirements. This cycle continues until the electric tricycle reaches the destination, ensuring the maximum energy efficiency of the electric tricycle during the entire driving process, while maintaining the smoothness and safety of driving.

[0030] Further, as Figure 2 shown, this application provides the electric vehicle controller to perform in-mode power distribution dynamic optimization and smooth inter-mode switching of the M power distribution modes corresponding to the connection sequence of the M local sections during the process of the electric tricycle traveling along the navigation route until the electric tricycle reaches the destination, including: When the electric tricycle enters the first local section of the navigation route, the power distribution is smoothly switched to the first power distribution mode through the electric vehicle controller, and the vehicle monitoring module is activated; the vehicle status data is obtained by real-time collection through the vehicle monitoring module, and power distribution dynamic optimization is performed in the first power distribution mode according to the vehicle status data until the electric tricycle enters the second local section of the navigation route; when the electric tricycle enters the second local section of the navigation route, the power distribution is smoothly switched from the first power distribution mode to the second power distribution mode through the electric vehicle controller, and power distribution dynamic optimization in the second power distribution mode is performed according to the updated vehicle status data of the vehicle monitoring module; and so on, performing in-mode power distribution dynamic optimization and smooth inter-mode switching of M - 2 power distribution modes between M - 2 local sections.

[0031] Preferably, when the electric tricycle enters the first partial section of the navigation route, the electric vehicle controller first calculates the median of each interval of the first power distribution mode, and smoothly switches the current power distribution to the first power distribution mode adapted to the conditions of this section according to the calculated median. While switching, the controller activates the vehicle monitoring module to start real-time collection of the operating state data of the electric tricycle, including battery state, driving state, and environmental state, etc. These data are dynamically monitored and fed back through the monitoring module as the dynamic optimization basis under the current power distribution mode; in the first partial section, the vehicle monitoring module continuously provides real-time state data, and the electric vehicle controller dynamically adjusts the output power, torque, and power distribution ratio according to these data in combination with the first power distribution adjustment model to ensure energy efficiency optimization and power adaptation under the first power distribution mode; when the electric tricycle is about to enter the second partial section, the vehicle monitoring module synchronously updates the state data, and the electric vehicle controller will smoothly switch the power distribution mode from the first power distribution mode to the second power distribution mode according to the road condition changes and the state of the electric tricycle. For example, when transitioning from a flat section to a steep slope section, it will smoothly switch to the load mode and adjust the output power and torque to meet the climbing requirements. During this process, the vehicle monitoring module continues to collect real-time data, and the controller uses the updated vehicle state data to dynamically optimize the second power distribution mode to ensure the energy efficiency and power matching of the electric tricycle when driving in the new section; similarly, cyclic operations will be performed between each partial section of the navigation route. When the electric tricycle enters the next section from the current partial section, the controller will smoothly switch between the modes according to the real-time state data and the road condition characteristics, and at the same time dynamically optimize the power distribution under the current mode. In this way, the power distribution mode switching and optimization of M - 2 partial sections are completed in sequence until the electric tricycle successfully reaches the final destination. The whole process not only ensures the efficient distribution of the power of the electric tricycle during driving, but also improves the smoothness and comfort of driving.

[0032] Furthermore, when the electric tricycle enters the first partial section of the navigation route, the present application provides a method of smoothly switching the power distribution to the first power distribution mode through the electric vehicle controller and activating the vehicle monitoring module, and the method includes: Call the first output power interval, the first torque interval, the first current interval, and the first power distribution ratio interval of the first power distribution mode from the K power distribution modes; extract the median of the first output power interval, the first torque interval, the first current interval, and the first power distribution ratio interval to obtain the first standard power parameter; when the electric tricycle enters the first partial section of the navigation route, smoothly switch the power distribution to the first standard power parameter through the electric vehicle controller and activate the vehicle monitoring module.

[0033] Optionally, first, call the first power distribution mode that matches the current local road section conditions from the pre-cured K power distribution modes, and obtain the key parameter intervals of this mode, including the first output power interval, the first torque interval, the first current interval, and the first power distribution ratio interval. These parameter intervals respectively define the applicable ranges of the power mode in terms of power output, torque, current, and power distribution ratio; after calling the above intervals, extract the median values of each parameter interval of the first power distribution mode, and calculate the first standard power parameters, including the first output power median, the first torque median, the first current median, and the first power distribution ratio median. These median parameters serve as the initial execution criteria for the first power distribution mode to ensure stable and efficient power distribution in the initial road section; when the electric tricycle enters the first local road section of the navigation route, smoothly switch the current power distribution to the calculated first standard power parameters through the electric vehicle controller. During the smooth switching process, sudden changes in power output will be avoided, and continuous transition of parameters will be achieved by gradually adjusting the output power, current, and torque values to ensure the smooth operation of the electric tricycle. At the same time, while the controller switches the power, it activates the vehicle monitoring module to start real-time collection of the state data of the electric tricycle (such as load, speed, battery status, etc.) to provide the necessary data support for subsequent dynamic optimization. This process ensures that the electric tricycle travels at the power parameters most suitable for the current road conditions and dynamic load in the initial local road section, laying a foundation for the power adjustment of subsequent local road sections.

[0034] Furthermore, this application provides for real-time collection of vehicle state data through the vehicle monitoring module, and dynamic optimization of power distribution in the first power distribution mode according to the vehicle state data until the electric tricycle enters the second local road section of the navigation route. The method includes: The vehicle status data is collected in real time through the vehicle monitoring module, where the vehicle status data includes real-time battery status data, real-time driving status data, and real-time environmental status data; constrained by the first output power range, first torque range, first current range, and first power distribution ratio range of the first power distribution mode, network data is called to obtain a composition of multiple sample battery status data, multiple sample driving status data, multiple sample environmental status data, and multiple sample power parameters, where the sample power parameter composition includes sample output power, sample torque, sample current, and sample power distribution ratio; the multiple sample battery status data, multiple sample driving status data, multiple sample environmental status data, and multiple sample power parameters are used as training data to construct a first power distribution adjustment model; the real-time battery status data, real-time driving status data, and real-time environmental status data are loaded into the first power distribution adjustment model to obtain updated power parameters; the updated power parameters are used to replace the first standard power parameters; a preset update period is set, and the first power distribution adjustment model is constrained by the update period to perform dynamic optimization of power distribution according to the information transmitted back by the vehicle monitoring module until the electric tricycle enters the second local section of the navigation route.

[0035] Optionally, the operation status data of the electric tricycle is collected in real time through the vehicle monitoring module, including real-time battery status data (such as the remaining percentage of battery power, instantaneous current and voltage), real-time driving status data (such as vehicle speed, load weight), and real-time environmental status data (such as road surface gradient, friction coefficient and traffic flow). These real-time data are used to evaluate the operation conditions of the electric tricycle on the current road section. On this basis, taking the parameter intervals of the first power distribution mode (such as the first output power interval, the first torque interval, the first current interval and the first power distribution ratio interval) as the constraint conditions, a large number of sample data under the corresponding conditions are called from the historical database or the remote cloud. These sample data include battery status data similar to the current operation conditions (such as historical voltage and current characteristics), driving status data (such as historical vehicle speed and load data), environmental status data (such as historical gradient and friction coefficient data), and the corresponding sample power parameters (including sample output power, sample torque, sample current and sample power distribution ratio). These sample data provide a sufficient data basis for constructing the first power distribution adjustment model. Using the collected sample data (multiple sample battery status data, multiple sample driving status data, multiple sample environmental status data and sample power parameters), the first power distribution adjustment model is constructed through a training algorithm. This model is used to establish the mapping relationship between the real-time status of the electric tricycle and the power distribution parameters. For example, when constructing the first power distribution adjustment model through a neural network, first use the neural network to construct the first power distribution adjustment model, including the input layer, the hidden layer and the output layer. Define the input layer according to the feature dimensions of the input data (such as the parameters of the battery status, driving status and environmental status), and the output layer corresponds to the power parameters (such as output power, torque, current and power distribution ratio). Subsequently, initialize the neural network weight and bias parameters. Methods such as Xavier initialization or He initialization can be used to ensure a reasonable initial weight distribution. Then, input the sample data (including sample battery status data, sample driving status data, sample environmental status data and sample power parameters) into the initialized neural network for forward propagation, passing through the input layer, the hidden layer, the activation function (such as ReLU) and the output layer in sequence, and calculating the predicted power parameter values.After that, the mean squared error (MSE) loss function is used to calculate the error value between the predicted results and the true sample power parameters. The gradients of the loss with respect to the network weights and biases are calculated layer by layer through the backpropagation algorithm. To accelerate optimization and avoid local optimum problems, the Adam optimizer is used to adjust the model parameters. The step size of the weight update is controlled by the learning rate to minimize the value of the loss function. The above training process is repeated, and batch sample data is input successively for training until the maximum number of iterations is reached or the loss function converges. After the training is completed, the model performance is tested, and the accuracy and robustness of the model in different power parameter prediction tasks are evaluated. If the predicted results meet the set accuracy requirements, the current neural network model is output as the first power distribution adjustment model. Otherwise, the hyperparameters of the model, such as the learning rate, the number of hidden layer units, the batch size, etc., are adjusted, and the model is retrained to further optimize the prediction performance. Through this process, an accurate and efficient first power distribution adjustment model is constructed to provide reliable support for real-time dynamic power optimization; during real-time operation, the real-time battery state data, real-time driving state data, and real-time environmental state data obtained through the vehicle monitoring module are loaded into the constructed first power distribution adjustment model. The model generates updated power parameters (such as updated output power, torque, current, and power distribution ratio) according to the input real-time state data. These updated parameters are used to replace the original first standard power parameters to adapt to the current dynamic operating conditions. At the same time, a preset update period is set to regularly obtain the feedback information from the vehicle monitoring module and input this information into the adjustment model for iterative optimization. This dynamic optimization process will continue until the electric tricycle enters the second local section of the navigation route; by continuously adjusting the power distribution parameters, it is ensured that the electric tricycle always adapts to the real-time operating environment with the best energy efficiency and performance, thereby improving the overall operating efficiency and driving stability of the electric tricycle.;

[0036] In summary, the embodiments of the present application have at least the following technical effects: By combining navigation route parsing, real-time load acquisition, and dynamic power distribution, the embodiments of the present application can generate multiple power distribution modes according to the real-time position, load status, and local road conditions of the electric tricycle, and dynamically adjust the output power, torque, current, and power distribution ratio through in-mode optimization and smooth switching between modes to meet the driving requirements of different sections; by constructing a mode database and a real-time power distribution adjustment model, and using historical data and real-time state data for matching analysis, it is ensured that the power distribution is efficient and accurate, realizing energy efficiency optimization and improvement of driving performance; these technical effects together solve the technical problems of low power distribution efficiency and insufficient endurance of electric vehicles under complex road conditions and dynamic loads, and achieve the effect of automatically adapting to different sections and load conditions through global road condition parsing and load perception, improving the power distribution efficiency and endurance of electric vehicles.

[0037] It should be noted that the above sequence of embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. And the above description of specific embodiments of this specification has been made. The processes depicted in the drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0038] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included within the protection scope of the present application.

[0039] This specification and the drawings are only exemplary descriptions of the present application and are considered to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. A real-time dynamic power distribution optimization method for an electric vehicle controller, characterized in that: The method comprises: After receiving the destination input by the target user, a navigation route is generated according to the real-time position of the electric tricycle and the destination; By performing global road condition analysis on the navigation route, M local road sections and M local road condition information are obtained, wherein the M local road sections are connected to restore the navigation route; The interactive load sensor obtains the real-time load of the electric tricycle; Perform power allocation demand analysis according to the real-time load and M local road condition information, and output M power allocation modes; When the electric tricycle is traveling along the navigation route, the electric vehicle controller performs dynamic optimization of power distribution within the M power distribution modes and smooth switching between modes according to the connection order of the M local sections until the electric tricycle reaches the destination.

2. A real-time dynamic power allocation optimization method for an electric vehicle controller as claimed in claim 1, characterized in that: The electric vehicle controller performs dynamic optimization of power allocation within the M power allocation modes and smooth switching between the modes according to the connection sequence of the M local sections during the driving of the electric tricycle along the navigation route until the electric tricycle reaches the destination. The method includes: When the electric tricycle enters the first local section of the navigation route, the power distribution is smoothly switched to the first power distribution mode through the electric vehicle controller, and the vehicle monitoring module is activated; The vehicle status data is acquired in real time by the vehicle monitoring module, and the power distribution is dynamically optimized in the first power distribution mode according to the vehicle status data until the electric tricycle enters the second local section of the navigation route; When the electric tricycle enters the second local section of the navigation route, the electric vehicle controller smoothly switches the power distribution from the first power distribution mode to the second power distribution mode, and dynamically optimizes the power distribution in the second power distribution mode according to the vehicle status data updated by the vehicle monitoring module; By analogy, dynamic optimization of power allocation within M-2 power allocation modes and smooth switching between modes are performed between M-2 local sections.

3. The real-time dynamic power distribution optimization method of an electric vehicle controller as claimed in claim 1, characterized in that: The interactive load sensor obtains the real-time load of the electric tricycle, and performs power distribution demand analysis according to the real-time load and M local road condition information, and outputs M power distribution modes. The method includes: Interactively obtain a sample load set, a sample road condition information set and a sample dynamic load set; Performing a multivariate regression analysis on the sample load set, the sample road condition information set and the sample dynamic load set, and constructing a dynamic load function based on the analysis results; Loading the real-time load and M local road condition information into the dynamic load function to perform dynamic load calculation and update, and obtaining M local dynamic loads; Pre-built schema database; The M local dynamic loads and the M local road condition information are loaded into the pattern database for real-time pattern matching, and the M power allocation patterns are output.

4. A real-time dynamic power distribution optimization method for an electric vehicle controller as claimed in claim 3, characterized in that: A pre-built pattern database, the method comprising: Solidify K power distribution modes, where each power distribution mode includes output power range, torque range, current range, power distribution ratio range and power priority setting; Calling a historical power log of the electric tricycle, wherein the historical power log records a plurality of historical dynamic loads, a plurality of historical road condition information, a plurality of historical power distribution information and a plurality of historical unit power consumptions, and the historical power distribution information includes an output power record, a torque record, a current record and a power distribution ratio record; According to the interval matching relationship between the multiple historical power allocation information and the K power allocation modes, the multiple historical dynamic loads, the multiple historical road condition information, and the multiple historical power allocation information are divided into groups to obtain K groups of historical dynamic loads, K groups of historical road condition information, and K groups of historical unit power consumption that are associated and mapped; Solving the power probability density of the K groups of historical unit power consumption to obtain K power consumption fluctuation intervals; Performing mode fitness analysis on the K groups of historical dynamic loads and K groups of historical road condition information according to the K power consumption fluctuation intervals to obtain K dynamic load intervals and K groups of road condition characteristic intervals; The K dynamic load intervals, K groups of road condition characteristic intervals and K power allocation modes are stored in association based on the knowledge graph to complete the construction of the mode database.

5. A real-time dynamic power distribution optimization method for an electric vehicle controller as claimed in claim 4, characterized in that: According to the K power consumption fluctuation intervals, mode fitness analysis is performed on the K groups of historical dynamic loads and K groups of historical road condition information to obtain K dynamic load intervals and K groups of road condition characteristic intervals, the method comprising: Reversely screening the K groups of historical dynamic loads and the K groups of historical road condition information according to the power consumption deviations of the K power consumption fluctuation intervals and the K groups of unit power consumption to obtain K groups of screened dynamic loads and K groups of screened road condition information; Calling load extreme values ​​on the K groups of screened dynamic loads to obtain the K dynamic load intervals; pre-define a power allocation associated characteristic set, wherein the power allocation associated characteristic set includes traffic flow characteristics, slope characteristics, elevation characteristics, and friction characteristics; The power allocation associated characteristic set is used to perform multi-index extreme value calls on the K groups of filtered road condition information to obtain the K groups of road condition characteristic intervals.

6. A real-time dynamic power distribution optimization method for an electric vehicle controller as claimed in claim 4, characterized in that: When the electric tricycle enters the first local section of the navigation route, the power distribution is smoothly switched to the first power distribution mode through the electric vehicle controller, and the vehicle monitoring module is activated. The method includes: Calling a first output power interval, a first torque interval, a first current interval and a first power distribution ratio interval of a first power distribution mode from the K power distribution modes; Performing median extraction on the first output power interval, the first torque interval, the first current interval, and the first power distribution ratio interval to obtain a first standard power parameter; When the electric tricycle enters the first local section of the navigation route, the power distribution is smoothly switched to the first standard power parameter through the electric vehicle controller, and the vehicle monitoring module is activated.

7. A real-time dynamic power distribution optimization method for an electric vehicle controller as claimed in claim 6, characterized in that: The vehicle status data is acquired in real time by the vehicle monitoring module, and the power distribution is dynamically optimized in the first power distribution mode according to the vehicle status data until the electric tricycle enters the second local section of the navigation route. The method includes: The vehicle status data is acquired in real time by the vehicle monitoring module, wherein the vehicle status data includes real-time battery status data, real-time driving status data and real-time environment status data; Taking the first output power interval, the first torque interval, the first current interval and the first power distribution ratio interval of the first power distribution mode as constraints, network data call is performed to obtain a plurality of sample battery status data, a plurality of sample driving status data, a plurality of sample environmental status data and a plurality of sample power parameter compositions, wherein the sample power parameter composition includes a sample output power, a sample torque, a sample current and a sample power distribution ratio; Using the plurality of sample battery state data, the plurality of sample driving state data, the plurality of sample environmental state data and the plurality of sample power parameters as training data, to construct a first power allocation adjustment model; Loading the real-time battery status data, the real-time driving status data and the real-time environment status data into the first power allocation adjustment model to obtain updated power parameters; Replacing the first standard power parameter with the updated power parameter; An update cycle is preset, and the first power allocation adjustment model is constrained by the update cycle, and dynamically optimizes power allocation according to the information returned by the vehicle monitoring module until the electric tricycle enters the second local section of the navigation route.

8. The real-time dynamic power distribution optimization method of an electric vehicle controller as claimed in claim 5, characterized in that: By performing global traffic analysis on the navigation route, M local road sections and M local traffic information are obtained, and the method includes: Obtaining global traffic information of the navigation route through a map API call; By traversing the global road condition information using the power allocation associated characteristic set, a plurality of groups of associated characteristic change nodes are located on the navigation route; Fine-grained splitting of the navigation route according to the multiple groups of associated characteristic change nodes to obtain the M local sections; The global traffic condition information is decomposed with the M local road sections as constraints to obtain M local traffic condition information.

Images