A current prediction control method for automotive rectifier

Abstract

The present invention discloses a current prediction control method for automobile rectifiers, which relates to the technical field of current prediction control. The method comprises: extracting a target rectifier according to an automobile component module; interacting with an Internet of Things monitoring module of a target automobile to obtain a rectifier monitoring data set of the target rectifier; receiving a rectifier current prediction instruction according to the automobile interaction module; activating a multi-coupling current prediction module according to the rectifier current prediction instruction; based on the rectifier monitoring data set, the expected current prediction window and the expected current prediction result, mining current prediction features according to the multi-coupling current prediction module to generate a current prediction variation signal; activating an adaptive control module to control the target rectifier according to the current prediction variation signal. The present invention solves the technical problems of low control accuracy and poor adaptability in the current control of automobile rectifiers in the prior art, and achieves the technical effect of improving the performance and safety of the automobile.

Description

Technical Field

[0001] The invention relates to the technical field of current prediction control, and in particular to a current prediction control method for an automobile rectifier. Background Art

[0002] With the rapid development of the automobile industry, the performance stability and current control accuracy of automobile rectifiers, as an important part of the automobile electrical system, are crucial to ensure the safe operation and efficient performance of the automobile. However, in the current automobile manufacturing and operation process, the current control of rectifiers faces many challenges.

[0003] First, with the increasing number of automobile functions and the increasing complexity of electrical systems, the requirements for rectifier current control are becoming higher and higher. Traditional current control methods often make it difficult to accurately predict and control rectifier current, which may lead to current fluctuations, overheating, and even damage during the operation of the car, seriously affecting the safety and reliability of the car.

[0004] Secondly, the current demand of the rectifier will also change under different working conditions of the car. For example, when accelerating, climbing a hill or driving at high speed, the rectifier needs to provide more current to meet the needs of the engine and electrical equipment. However, traditional current control methods often cannot adapt to these changes in real time, resulting in insufficient or excessive current supply, affecting the performance and energy consumption of the car. Summary of the invention

[0005] The present application provides a current prediction control method for an automobile rectifier, which is used to solve the technical problems of low control accuracy and poor adaptability in the prior art for automobile rectifier current control.

[0006] In view of the above problems, the present application provides a current prediction control method for an automotive rectifier.

[0007] In a first aspect of the present application, a current prediction control method for an automotive rectifier is provided, the method comprising:

[0008] According to the automobile constituent module, a target rectifier is extracted, wherein the automobile constituent module includes multiple automobile components of the target automobile, and the target rectifier is any automobile rectifier of the target automobile; the Internet of Things monitoring module of the target automobile is interacted with to obtain a rectifier monitoring data set of the target rectifier; according to the automobile interaction module, a rectifier current prediction instruction is received, wherein the rectifier current prediction instruction includes an expected current prediction window, and an expected current prediction result corresponding to the expected current prediction window; according to the rectifier current prediction instruction, a multi-coupling current prediction module is activated, wherein the multi-coupling current prediction module includes a multilateral current prediction composite model, a current prediction feature recognition model and a current prediction verification model; based on the rectifier monitoring data set, the expected current prediction window and the expected current prediction result, current prediction feature mining is performed according to the multi-coupling current prediction module to generate a current prediction variation signal; according to the current prediction variation signal, an adaptive control module is activated to control the target rectifier, wherein the adaptive control module includes an optimal control decision submodule and an automobile control submodule.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] The present application extracts a target rectifier according to a vehicle constituent module, wherein the vehicle constituent module includes multiple vehicle components of the target vehicle, and the target rectifier is any vehicle rectifier of the target vehicle; interacts with the Internet of Things monitoring module of the target vehicle to obtain a rectifier monitoring data set of the target rectifier; receives a rectifier current prediction instruction according to the vehicle interaction module, wherein the rectifier current prediction instruction includes an expected current prediction window, and an expected current prediction result corresponding to the expected current prediction window; activates a multi-coupled current prediction module according to the rectifier current prediction instruction, wherein the multi-coupled current prediction module includes a multilateral current prediction composite model, a current prediction feature recognition model and a current prediction verification model; based on the rectifier monitoring data set, the expected current prediction window and the expected current prediction result, current prediction feature mining is performed according to the multi-coupled current prediction module to generate a current prediction variation signal; according to the current prediction variation signal, an adaptive control module is activated to control the target rectifier, wherein the adaptive control module includes an optimal control decision submodule and a vehicle control submodule. The present invention solves the technical problems of low control accuracy and poor adaptability in the prior art for automobile rectifier current control, and achieves the technical effect of improving the performance and safety of the automobile by accurately predicting and controlling the current. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0012] Figure 1 A schematic flow chart of a current prediction control method for an automobile rectifier provided in an embodiment of the present application;

[0013] Figure 2 A schematic diagram of a flow chart of generating a current prediction variation signal in a current prediction control method for an automotive rectifier provided in an embodiment of the present application. DETAILED DESCRIPTION

[0014] The present application provides a current prediction and control method for an automobile rectifier, aiming to solve the technical problems of low control accuracy and poor adaptability in the prior art for automobile rectifier current control, and achieves the technical effect of improving the performance and safety of the automobile through accurate prediction and control of the current.

[0015] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0016] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices.

[0017] Embodiment 1

[0018] like Figure 1 As shown, the present application provides a current prediction control method for an automotive rectifier, the method comprising:

[0019] Step S100: extracting a target rectifier according to a vehicle component module, wherein the vehicle component module includes a plurality of vehicle components of a target vehicle, and the target rectifier is any one of the vehicle rectifiers of the target vehicle;

[0020] In the embodiment of the present application, the automobile component module includes various components of the target automobile, such as the engine, transmission system, electrical system and other parts. Among the components of the automobile component module, the rectifier is a key component responsible for converting AC power into DC power and providing a stable DC power supply for the automobile's electrical system. Since there are multiple rectifiers in the automobile, one is selected from the numerous rectifiers as the target rectifier according to actual needs.

[0021] Step S200: interacting with the Internet of Things monitoring module of the target vehicle to obtain a rectifier monitoring data set of the target rectifier;

[0022] In the embodiment of the present application, the IoT monitoring module is used to monitor and collect the operating data of each component of the target vehicle in real time. The IoT monitoring module includes sensors, data acquisition equipment, and data transmission networks, which sense and record the operating status, performance parameters, and environmental information of each component of the vehicle in real time.

[0023] Before interacting with the IoT monitoring module of the target vehicle, the identification information of the target sensor is determined according to the extracted target rectifier, such as the rectifier model, serial number, etc. After the interaction, the IoT monitoring module selects data related to the target rectifier from the monitoring data, including various parameters such as the rectifier's input voltage, output voltage, output current, and temperature, according to the identification information.

[0024] Finally, the filtered data are integrated to obtain the rectifier monitoring data set of the target rectifier.

[0025] Step S300: receiving a rectifier current prediction instruction according to the automobile interaction module, wherein the rectifier current prediction instruction includes an expected current prediction window and an expected current prediction result corresponding to the expected current prediction window;

[0026] In the embodiment of the present application, the vehicle interaction module realizes information interaction between the vehicle and the user and various systems inside the vehicle. When performing rectifier current prediction, the user or the system sends a rectifier current prediction instruction to the vehicle interaction module through touch screen operation, voice command, etc. The rectifier current prediction instruction includes an expected current prediction window and a corresponding expected current prediction result.

[0027] The expected current prediction window refers to the time range in which the user or system hopes to predict the rectifier current. This time range is a few minutes, a few hours, or longer. The expected current prediction result within the corresponding window is the rectifier current value or current range that the user or system expects to reach within the prediction window. The expected result is determined based on multiple factors such as the operating status of the car, the driving environment, and user needs. For example, when driving at high speed, a higher current output is required to ensure power performance; while when driving at low speed or idling, a lower current output is required to save energy.

[0028] Step S400: activating a multi-coupling current prediction module according to the rectifier current prediction instruction, wherein the multi-coupling current prediction module includes a multilateral current prediction composite model, a current prediction feature recognition model and a current prediction verification model;

[0029] In the embodiment of the present application, after receiving the rectifier current prediction instruction, the multi-coupled current prediction module is activated to perform current prediction according to the expected current prediction window and the expected current prediction result in the instruction.

[0030] The multi-coupling current prediction module includes three sub-models, namely, a multi-lateral current prediction composite model, a current prediction feature recognition model, and a current prediction verification model. These three sub-models work together to complete the current prediction task.

[0031] The multilateral current prediction composite model combines a variety of current prediction algorithms and techniques, such as time series analysis, machine learning, neural networks, etc., to achieve multivariate prediction of rectifier current.

[0032] The current prediction feature recognition model analyzes and processes the prediction results. Using feature extraction and recognition technology, key features related to current changes are extracted from the rectifier monitoring data. Key features include voltage fluctuations, temperature changes, load changes, etc.

[0033] The current prediction verification model verifies and corrects the prediction results. The prediction results are verified and evaluated based on historical data, real-time data and other relevant information. If there is a large deviation between the prediction results and the actual data, the current prediction verification model adjusts and optimizes it to improve the accuracy and reliability of the prediction.

[0034] Step S500: Based on the rectifier monitoring data set, the expected current prediction window and the expected current prediction result, current prediction feature mining is performed according to the multi-coupling current prediction module to generate a current prediction variation signal;

[0035] In the embodiment of the present application, the rectifier monitoring data set is imported into the multi-coupling current prediction module as input data, and the expected current prediction window is set. The multi-coupling current prediction module performs current prediction calculation based on the input data and historical data to obtain the predicted current value or current range within the expected current prediction window.

[0036] Then, the predicted current value is compared with the expected current prediction result by calculating the deviation, variance and other statistics between the predicted value and the expected value. If there is a significant difference between the predicted current value and the expected current prediction result, which exceeds the preset threshold range, a current prediction variation signal is generated. The threshold range is set according to the actual situation, for example, set to ±5% or ±10% of the expected value.

[0037] Step S600: activating an adaptive control module to control the target rectifier according to the current prediction variation signal, wherein the adaptive control module includes an optimal control decision submodule and a vehicle control submodule.

[0038] In the embodiment of the present application, when the current prediction variation signal is received, the adaptive control module starts and initializes the relevant submodules. The optimal control decision submodule starts to analyze the characteristics and trends of the current prediction variation signal and determines the appropriate control strategy. The optimal control decision submodule controls the parameter combination according to methods such as genetic algorithms and particle swarm optimization, and formulates a control strategy that meets actual needs according to the overall operating status and load conditions of the vehicle.

[0039] After determining the control strategy, the vehicle control submodule generates corresponding control instructions according to the control strategy and sends them to the target rectifier through the vehicle control system. The control instructions include adjusting the voltage and current output of the rectifier, changing its working mode, etc. Through precise control, the vehicle control submodule can achieve real-time adjustment and optimization of the rectifier performance.

[0040] During the control process, the adaptive control module continuously monitors and evaluates the operating status and performance of the rectifier. It collects real-time data from the rectifier and compares it with the expected current prediction results. If it is found that there are still current prediction variation signals or poor performance, the adaptive control module readjusts the control strategy and executes the control process again until the expected current output and performance requirements are achieved.

[0041] Further, such as Figure 2 As shown, step S500 in the method provided in the application embodiment also includes:

[0042] Based on the rectifier monitoring data set and the expected current prediction window, current prediction is performed according to the multilateral current prediction composite model to obtain a rectifier current prediction result;

[0043] Inputting the rectifier current prediction result and the expected current prediction result into the current prediction feature recognition model to generate a current prediction variation coefficient;

[0044] Inputting the current prediction variation coefficient into the current prediction verification model, wherein the current prediction verification model includes a current prediction variation threshold;

[0045] If the current prediction variation coefficient is greater than / equal to the current prediction variation threshold, the current prediction variation signal is obtained.

[0046] In an embodiment of the present application, historical data related to current prediction is extracted from a rectifier monitoring data set. These data include real-time measurement values ​​or time series data of parameters such as voltage, current, temperature, and load. According to the expected current prediction window, the time range of current prediction is determined. The rectifier monitoring data set is input into a multilateral current prediction composite model, and the multilateral current prediction composite model performs current prediction calculations based on the input historical data and the expected current prediction window, and obtains the rectifier current prediction result within the expected current prediction window through calculation.

[0047] The obtained rectifier current prediction results and expected current prediction results are simultaneously input into the current prediction feature recognition model. The current prediction feature recognition model performs difference analysis and feature extraction based on the input prediction results and expected results. The difference between the prediction results and the expected values ​​is quantified by calculating the deviation, relative error or other statistics between the predicted value and the expected value. Based on the results of the difference analysis and feature extraction, the current prediction coefficient of variation is generated.

[0048] The generated current prediction coefficient of variation is input into the current prediction verification model. The current prediction verification model contains one or more current prediction variation thresholds, which are set according to historical data, system requirements or expert experience to determine the reliability of the current prediction results. If the current prediction coefficient of variation is greater than or equal to the current prediction variation threshold, it means that the difference between the predicted current and the expected current exceeds the acceptable range. At this time, the current prediction verification model will output a current prediction variation signal. The current prediction variation signal indicates that there is a significant variation in the current prediction results.

[0049] Further, based on the rectifier monitoring data set and the expected current prediction window, current prediction is performed according to the multilateral current prediction composite model to obtain a rectifier current prediction result, and the method also includes:

[0050] The multi-lateral current prediction composite model includes a first rectifier current prediction model and a second rectifier current prediction model, wherein the first rectifier current prediction model is used to predict a rectifier current less than or equal to a first preset window, the second rectifier current prediction model is used to predict a rectifier current greater than or equal to a second preset window, and the first preset window is smaller than the second preset window;

[0051] If the expected current prediction window is less than or equal to the first preset window, the first rectifier current prediction model is activated to perform current prediction on the rectifier monitoring data set to obtain the rectifier current prediction result.

[0052] In an embodiment of the present application, the multilateral current prediction composite model includes a first rectifier current prediction model and a second rectifier current prediction model. The first rectifier current prediction model is used to predict the rectifier current less than or equal to the first preset window. The first preset window represents a shorter time range and is suitable for accurate prediction of recent or real-time current. The second rectifier current prediction model is used to predict the rectifier current greater than or equal to the second preset window. The second preset window is larger than the first preset window and is suitable for predicting the current trend over a longer period of time.

[0053] Before making a current prediction, determine the size of the expected current prediction window. If the expected current prediction window is less than or equal to the first preset window, use the first rectifier current prediction model for prediction. If the expected current prediction window is greater than the first preset window but less than or equal to the second preset window, select a model or adopt a combination strategy for prediction according to the specific situation. If the expected current prediction window is greater than the second preset window, use the second rectifier current prediction model for prediction.

[0054] When the expected current prediction window is less than or equal to the first preset window, the first rectifier current prediction model is activated. Historical data corresponding to the expected current prediction window is extracted from the rectifier monitoring data set. These data are input into the first rectifier current prediction model, and the model calculates and outputs a prediction result based on the input data, and the prediction result represents the rectifier current prediction value within the window of the expected current.

[0055] Furthermore, the method further comprises:

[0056] If the expected current prediction window is greater than or equal to the second preset window, the second rectifier current prediction model is activated to perform current prediction on the rectifier monitoring data set to obtain the rectifier current prediction result.

[0057] In an embodiment of the present application, when the expected current prediction window is greater than or equal to the second preset window, the second rectifier current prediction model is activated to perform current prediction on the rectifier monitoring data set.

[0058] After activating the second rectifier current prediction model, appropriate historical data are extracted from the rectifier monitoring data set as input to the model. These data cover the time period corresponding to the expected current prediction window, and some preprocessing operations such as data cleaning, normalization or feature extraction are performed to ensure the quality and consistency of the input data.

[0059] The prepared input data is then input into the second rectifier current prediction model. The model will calculate the prediction result based on the input historical data and its internal learning mechanism. After the calculation is completed, the rectifier current prediction result is output.

[0060] Furthermore, the method further comprises:

[0061] If the expected current prediction window is larger than the first preset window, and the expected current prediction window is smaller than the second preset window, generating a joint prediction instruction;

[0062] According to the joint prediction instruction, based on the rectifier monitoring data set, calling the first rectifier current prediction model and the second rectifier current prediction model to perform current prediction respectively, and generating a first current prediction result and a second current prediction result;

[0063] The first current prediction result and the second current prediction result are used to perform data fusion to obtain the rectifier current prediction result.

[0064] In the embodiment of the present application, when the expected current prediction window is between the first preset window and the second preset window, a joint prediction instruction is automatically generated.

[0065] According to the joint prediction instruction, the first rectifier current prediction model and the second rectifier current prediction model are called to perform current prediction based on the rectifier monitoring data set respectively. The first current prediction result and the second current prediction result are obtained through current prediction.

[0066] Different weights are assigned to the prediction results of the first rectifier current prediction model and the second rectifier current prediction model. The weights are assigned based on historical experience. The first prediction window is compared with the second prediction window to assign weights, and then weighted calculation is performed to obtain the rectifier current prediction result.

[0067] Furthermore, step S600 in the method provided in the application embodiment also includes:

[0068] The optimal control decision submodule includes an optimal control decision space and a control decision prediction and evaluation model;

[0069] Extracting a first rectifier control decision according to the optimal control decision space;

[0070] Performing a simulated control evaluation on the first rectifier control decision according to the control decision prediction and evaluation model to obtain a first simulated control evaluation matrix;

[0071] Determining whether the first simulated control evaluation matrix satisfies a preset control evaluation standard matrix;

[0072] If the first simulated control evaluation matrix satisfies the preset control evaluation standard matrix, outputting the first rectifier control decision as a rectifier control scheme;

[0073] The rectifier control scheme is encrypted and transmitted to the vehicle control submodule.

[0074] In an embodiment of the present application, the optimal control decision submodule includes an optimal control decision space and a control decision prediction and evaluation model.

[0075] The optimal control decision space contains a variety of possible rectifier control decision sets. When constructing the optimal control decision space, the experience and knowledge of domain experts are used to build the initial framework of the decision space. Then, the genetic algorithm is used to continuously optimize and expand the decision space to find more control decisions.

[0076] A number of groups of control decisions are randomly selected from the optimal control decision space as candidates. The candidate control decisions are prioritized according to historical data, and the highest priority is selected as the first rectifier control decision.

[0077] Then, the control decision prediction and evaluation model is used to simulate the control decision of the first rectifier. The control decision prediction and evaluation model simulates the operating status of the rectifier under a specific control decision through a neural network model and predicts its performance.

[0078] Specifically, first collect the historical operating data of the rectifier under different control decisions, including key performance indicators such as current, voltage, and temperature. Clean and preprocess the collected data, remove outliers, fill missing values, and standardize or normalize them. Then, according to the nature of the problem and the characteristics of the data, select the recurrent neural network as the prediction model. Then determine the input, output, and hidden layer structure of the model. The input includes the control decision parameters and the initial state of the rectifier, and the output is the predicted performance indicator. The number of hidden layers and nodes is adjusted according to the complexity of the problem and the performance of the model.

[0079] The preprocessed data is divided into training set, validation set and test set. The training set is used to train the model, the validation set is used to adjust the model's hyperparameters, and the test set is used to evaluate the model's performance. According to the nature of the prediction problem, select a suitable loss function, such as mean square error, cross entropy, etc., to measure the difference between the model's predicted value and the true value. At the same time, select a suitable optimization algorithm, such as gradient descent, Adam, etc. to minimize the loss function. Use the training set to iteratively train the model. In each iteration, calculate the loss between the model's predicted value and the actual value, and update the model's parameters through the optimization algorithm to reduce the loss. Repeat this process until the model's performance on the validation set is stable or meets the preset stop condition. Finally, use the test set to evaluate the performance of the trained model. Use grid search, random search, etc. to adjust the model's hyperparameters to further optimize the model's performance. Finally, a trained regulatory decision prediction and evaluation model is obtained.

[0080] The control decision of the first rectifier is simulated and evaluated by the control decision prediction and evaluation model to generate a first simulated control evaluation matrix, which includes predicted values ​​of various performance indicators, such as rectification efficiency, current stability, temperature distribution, etc.

[0081] The first simulated control evaluation matrix is ​​compared and analyzed with the preset control evaluation standard matrix. The preset control evaluation standard matrix is ​​formulated according to factors such as system requirements, performance requirements, and safety restrictions to judge the quality of the simulated control evaluation results.

[0082] If the first simulated control evaluation matrix meets the preset control evaluation standard matrix, it means that the group of control decisions is effective and can achieve the expected performance target. At this time, the optimal control decision submodule outputs the first rectifier control decision as the final rectifier control plan.

[0083] The rectifier control scheme is encrypted using symmetric encryption algorithms and asymmetric encryption algorithms to generate ciphertext, and then the encrypted rectifier control scheme is transmitted to the vehicle control submodule through the vehicle network.

[0084] Further, a simulated control evaluation is performed on the first rectifier control decision according to the control decision prediction evaluation model to obtain a first simulated control evaluation matrix, and the method further includes:

[0085] The control decision prediction and evaluation model includes a control decision prediction model and a control prediction and evaluation model;

[0086] Based on the target rectifier, a simulated control of the first rectifier control decision is performed according to the control decision prediction model to generate a first simulated control result, wherein the control decision prediction model includes a multi-dimensional control decision prediction index, and the multi-dimensional control decision prediction index includes a simulated control current, a simulated control response speed, a simulated control loss and a simulated control fault;

[0087] Evaluate the first simulation control result according to the control prediction and evaluation model to generate a first control prediction and evaluation result, wherein the first control prediction and evaluation result includes a first current adaptation prediction coefficient, a first control response prediction coefficient, a first control loss prediction coefficient, and a first control fault prediction coefficient;

[0088] The first simulation regulation and evaluation matrix is ​​constructed by performing standardization processing based on the first regulation and control prediction evaluation result.

[0089] In the embodiment of the present application, the control decision prediction and evaluation model includes two parts: a control decision prediction model and a control prediction and evaluation model. The control decision prediction model is used to simulate the operating performance of the rectifier under different control decisions, and the control prediction and evaluation model evaluates these simulation results and generates corresponding evaluation coefficients.

[0090] The first rectifier control decision is input into the control decision prediction model, and the control decision prediction model simulates the control of the target rectifier according to the first rectifier control decision, and outputs the first simulated control result after the control is completed. The first simulated control result includes multi-dimensional control decision prediction indicators, such as simulated control current, simulated control response speed, simulated control loss and simulated control fault.

[0091] Afterwards, the control prediction evaluation model receives the first simulated control result output by the control decision prediction model as input. According to the preset evaluation criteria and methods, each indicator in the first simulated control result is evaluated. The evaluation process is a comparison with the ideal performance or historical data. After the evaluation, the control prediction evaluation model outputs the first control prediction evaluation result. The first control prediction evaluation result includes the first current adaptation prediction coefficient, the first control response prediction coefficient, the first control loss prediction coefficient, and the first control fault prediction coefficient.

[0092] The first control prediction evaluation results are standardized, and the values ​​of each evaluation index are made to fall within the same numerical range through data normalization or standardization conversion. After the standardization, each evaluation coefficient is organized in a certain format to construct the first simulation control evaluation matrix.

[0093] Furthermore, the method further comprises:

[0094] Performing fault prediction according to the rectifier monitoring data set to obtain a rectifier fault prediction result;

[0095] Performing fault evaluation according to the rectifier fault prediction result to generate a rectifier fault evaluation coefficient;

[0096] Determining whether the rectifier fault evaluation coefficient is less than a fault evaluation threshold;

[0097] If the rectifier fault evaluation coefficient is greater than / equal to the fault evaluation threshold, activating a pre-built rectifier fault operation and maintenance map;

[0098] Inputting the rectifier fault prediction result into the rectifier fault operation and maintenance map to obtain a matching fault operation and maintenance decision;

[0099] The rectifier control scheme is optimized according to the matching fault operation and maintenance decision, an optimized control scheme is generated, and the optimized control scheme is encrypted and transmitted to the vehicle control submodule.

[0100] In the embodiment of the present application, when fault prediction is performed based on the rectifier monitoring data set, the monitoring data is first cleaned and standardized to remove noise and outliers. Fault-related features are then extracted from the monitoring data, and the prediction model is trained using a neural network using historical fault data and corresponding features. The real-time monitoring data is input into the trained model to obtain the rectifier fault prediction result.

[0101] Analyze the rectifier fault prediction results to obtain information such as the fault type, fault probability or risk level of the prediction results. Determine the fault evaluation index based on the characteristics and application scenarios of the rectifier. The fault evaluation index includes fault type, fault probability, safety impact, etc. Use appropriate quantitative methods based on the fault evaluation index, such as calculating the rectifier fault evaluation coefficient through weight distribution. The weight is determined by analyzing historical data. Finally, the rectifier fault evaluation coefficient is generated by calculation. The rectifier fault evaluation coefficient is a value or a set of values ​​that reflects the severity and urgency of the fault.

[0102] The fault evaluation threshold is set based on historical data and experience. The rectifier fault evaluation coefficient is compared with the fault evaluation threshold to determine whether the rectifier fault evaluation coefficient is less than the fault evaluation threshold.

[0103] If the fault evaluation coefficient is greater than or equal to the set threshold, it means that the rectifier has a serious fault risk. At this time, the pre-built rectifier fault operation and maintenance map is activated. The rectifier fault operation and maintenance map is constructed by collecting and organizing rectifier fault cases, maintenance experience and best practices. The rectifier fault operation and maintenance map contains a database or knowledge base of fault types, causes, impacts and corresponding operation and maintenance strategies.

[0104] The fault prediction results of the rectifier are input into the activated fault operation and maintenance map, and the operation and maintenance decision that best matches the current fault situation is found through a matching algorithm, such as similarity-based matching. Based on the obtained matching fault operation and maintenance decision, the original rectifier control plan is optimized and adjusted to generate an optimized control plan.

[0105] Finally, the generated optimized control scheme is encrypted to ensure the security of data transmission, and then the encrypted control scheme is transmitted to the vehicle control sub-module.

[0106] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects:

[0107] The present application extracts a target rectifier according to a vehicle constituent module, wherein the vehicle constituent module includes multiple vehicle components of the target vehicle, and the target rectifier is any vehicle rectifier of the target vehicle; interacts with the Internet of Things monitoring module of the target vehicle to obtain a rectifier monitoring data set of the target rectifier; receives a rectifier current prediction instruction according to the vehicle interaction module, wherein the rectifier current prediction instruction includes an expected current prediction window, and an expected current prediction result corresponding to the expected current prediction window; activates a multi-coupled current prediction module according to the rectifier current prediction instruction, wherein the multi-coupled current prediction module includes a multilateral current prediction composite model, a current prediction feature recognition model and a current prediction verification model; based on the rectifier monitoring data set, the expected current prediction window and the expected current prediction result, current prediction feature mining is performed according to the multi-coupled current prediction module to generate a current prediction variation signal; according to the current prediction variation signal, an adaptive control module is activated to control the target rectifier, wherein the adaptive control module includes an optimal control decision submodule and a vehicle control submodule. The present invention solves the technical problems of low control accuracy and poor adaptability in the prior art for automobile rectifier current control, and achieves the technical effect of improving the performance and safety of the automobile by accurately predicting and controlling the current.

[0108] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. Other embodiments are within the scope of the attached claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0109] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0110] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations 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 equivalents, the present application intends to include these modifications and variations.

Claims

1. A current prediction control method for an automotive rectifier, characterized in that: The method comprises: Extracting a target rectifier according to a vehicle constituent module, wherein the vehicle constituent module includes a plurality of vehicle components of the target vehicle, and the target rectifier is any vehicle rectifier of the target vehicle; Interact with the IoT monitoring module of the target vehicle to obtain a rectifier monitoring data set of the target rectifier; According to the vehicle interaction module, a rectifier current prediction instruction is received, wherein the rectifier current prediction instruction includes an expected current prediction window and an expected current prediction result corresponding to the expected current prediction window; According to the rectifier current prediction instruction, a multi-coupling current prediction module is activated, wherein the multi-coupling current prediction module includes a multilateral current prediction composite model, a current prediction feature recognition model and a current prediction verification model; Based on the rectifier monitoring data set, the expected current prediction window and the expected current prediction result, the current prediction feature mining is performed according to the multi-coupling current prediction module to generate the current prediction variation signal; According to the current prediction variation signal, the adaptive control module is activated to control the target rectifier, wherein the adaptive control module includes an optimal control decision submodule and a vehicle control submodule; Based on the rectifier monitoring data set and the expected current prediction window, current prediction is performed according to the multilateral current prediction composite model to obtain the rectifier current prediction result; Inputting the rectifier current prediction result and the expected current prediction result into the current prediction feature recognition model to generate the current prediction variation coefficient; Inputting the current prediction variation coefficient into a current prediction verification model, wherein the current prediction verification model includes a current prediction variation threshold; If the current prediction variation coefficient is greater than / equal to the current prediction variation threshold, a current prediction variation signal is obtained.

2. The method according to claim 1, characterized in that Based on the rectifier monitoring data set and the expected current prediction window, current prediction is performed according to the multilateral current prediction composite model to obtain a rectifier current prediction result, including: The multi-lateral current prediction composite model includes a first rectifier current prediction model and a second rectifier current prediction model, wherein the first rectifier current prediction model is used to predict a rectifier current less than or equal to a first preset window, the second rectifier current prediction model is used to predict a rectifier current greater than or equal to a second preset window, and the first preset window is smaller than the second preset window; If the expected current prediction window is less than or equal to the first preset window, the first rectifier current prediction model is activated to perform current prediction on the rectifier monitoring data set to obtain the rectifier current prediction result.

3. The method according to claim 2, characterized in that If the expected current prediction window is greater than or equal to the second preset window, the second rectifier current prediction model is activated to perform current prediction on the rectifier monitoring data set to obtain the rectifier current prediction result.

4. The method according to claim 2, characterized in that The method comprises: If the expected current prediction window is larger than the first preset window, and the expected current prediction window is smaller than the second preset window, generating a joint prediction instruction; According to the joint prediction instruction, based on the rectifier monitoring data set, calling the first rectifier current prediction model and the second rectifier current prediction model to perform current prediction respectively, and generating a first current prediction result and a second current prediction result; The first current prediction result and the second current prediction result are data fused to obtain the rectifier current prediction result.

5. The method according to claim 1, characterized in that According to the current prediction variation signal, the adaptive control module is activated to control the target rectifier, wherein the adaptive control module includes an optimal control decision submodule and a vehicle control submodule, including: The optimal control decision submodule includes an optimal control decision space and a control decision prediction and evaluation model; Extracting a first rectifier control decision according to the optimal control decision space; Performing a simulated control evaluation on the first rectifier control decision according to the control decision prediction and evaluation model to obtain a first simulated control evaluation matrix; Determining whether the first simulated control evaluation matrix satisfies a preset control evaluation standard matrix; If the first simulated control evaluation matrix satisfies the preset control evaluation standard matrix, outputting the first rectifier control decision as a rectifier control scheme; The rectifier control scheme is encrypted and transmitted to the vehicle control submodule.

6. The method according to claim 5, characterized in that Performing a simulated control evaluation on the first rectifier control decision according to the control decision prediction and evaluation model to obtain a first simulated control evaluation matrix includes: The control decision prediction and evaluation model includes a control decision prediction model and a control prediction and evaluation model; Based on the target rectifier, a simulated control of the first rectifier control decision is performed according to the control decision prediction model to generate a first simulated control result, wherein the control decision prediction model includes a multi-dimensional control decision prediction index, and the multi-dimensional control decision prediction index includes a simulated control current, a simulated control response speed, a simulated control loss and a simulated control fault; Evaluate the first simulation control result according to the control prediction and evaluation model to generate a first control prediction and evaluation result, wherein the first control prediction and evaluation result includes a first current adaptation prediction coefficient, a first control response prediction coefficient, a first control loss prediction coefficient, and a first control fault prediction coefficient; The first simulation regulation and evaluation matrix is ​​constructed by performing standardization processing based on the first regulation and control prediction evaluation result.

7. The method according to claim 5, characterized in that The method comprises: Performing fault prediction according to the rectifier monitoring data set to obtain a rectifier fault prediction result; Performing fault evaluation according to the rectifier fault prediction result to generate a rectifier fault evaluation coefficient; Determining whether the rectifier fault evaluation coefficient is less than a fault evaluation threshold; If the rectifier fault evaluation coefficient is greater than / equal to the fault evaluation threshold, activating a pre-built rectifier fault operation and maintenance map; Inputting the rectifier fault prediction result into the rectifier fault operation and maintenance map to obtain a matching fault operation and maintenance decision; The rectifier control scheme is optimized according to the matching fault operation and maintenance decision, an optimized control scheme is generated, and the optimized control scheme is encrypted and transmitted to the vehicle control submodule.

Images