Fan control method and device for vehicle engine, engine and automobile
By acquiring historical vehicle data and using predictive models to optimize fan speed control, the problems of slow fan control response and high power consumption were solved, achieving precise adjustment of engine coolant temperature and improved fuel economy.
Patent Information
- Application Number
- CN202511381156.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-25
- Publication Date
- 2025-11-21
AI Technical Summary
In existing technologies, fan control has a slow response and high power consumption, which cannot effectively regulate engine coolant temperature, leading to engine failure or poor fuel economy.
By acquiring historical vehicle data and using predictive models to forecast future vehicle data, and combining this with preset control functions to optimize fan speed control, advance adjustments can be achieved.
Reduce fan control response time, improve water temperature regulation accuracy, reduce power consumption, and enhance engine operating stability and fuel economy.
Smart Images
Figure CN120990732A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the automotive field, specifically to a fan control method, device, engine, and automobile for a vehicle engine. Background Technology
[0002] The stability of a car engine's coolant temperature directly affects power output, fuel economy, and engine lifespan. When the engine is running, combustion in the combustion chamber and friction between mechanical components generate a significant amount of heat. If the coolant temperature is too high, it can easily lead to piston ring coking, cylinder block deformation, and even serious malfunctions such as cylinder scoring. Conversely, if the coolant temperature is too low, it will result in incomplete fuel combustion, increasing fuel consumption and emissions, while also accelerating engine wear. Therefore, precise coolant temperature control is a core objective of the engine's thermal management system.
[0003] As a key heat dissipation component in the engine cooling system, the efficiency of coolant temperature regulation is directly determined by the rationality of its power control. Currently, traditional automotive fan control calculates the target coolant temperature based on engine load, ambient temperature, and vehicle speed, and then calculates the target fan speed using the engine's target coolant temperature and the actual coolant temperature. This traditional control method cannot overcome the significant lag problem in temperature control. For example, when the coolant temperature rises to 98°C, the fan turns on, and the powerful airflow generated by the fan blows across the large surface area of the radiator's fins. The heat from the coolant inside the radiator pipes is efficiently conducted to the air through the pipe walls and fins and quickly carried away by the high-speed airflow. The delay in heat dissipation, the delay in coolant temperature sensor measurement, and the delay in fan response all contribute to high power consumption of the fan accessories. Summary of the Invention
[0004] In view of this, it is necessary to provide a fan control method, device, engine and automobile for a vehicle engine to solve the technical problems of greatly delayed fan control response and high power consumption in the prior art.
[0005] To achieve the aforementioned technical effects, firstly, this application provides a fan control method for a vehicle engine, comprising: Acquire historical vehicle data for a vehicle within a set historical time period; Based on the historical vehicle data, predict the vehicle data for a set future time period; The predicted engine coolant temperature is obtained based on the predicted vehicle data. The preset control function is solved based on the predicted water temperature to obtain the target speed of the fan. The fan is then controlled based on the target speed. The preset control function is the target function of the vehicle's engine water temperature and fan speed.
[0006] In one possible embodiment, predicting the vehicle data for a given future time period based on the historical vehicle data includes: The predicted vehicle data is obtained based on a preset correspondence and the historical vehicle data. The preset correspondence is the correspondence between different vehicle data obtained from the engine universal test.
[0007] By pre-determining the correspondence between data from different vehicles through universal engine testing, the correspondence between different vehicle data can be made more realistic, thereby making the predicted vehicle data more realistic and improving the accuracy of the final fan control.
[0008] In one possible embodiment, predicting the vehicle data for a given future time period based on the historical vehicle data includes: Obtain a vehicle data prediction model pre-trained based on sample vehicle data, input the historical vehicle data into the vehicle data prediction model, and obtain the output of the vehicle data prediction model as the predicted vehicle data.
[0009] In one possible embodiment, the vehicle data prediction model is a neural network model built on a long short-term memory network.
[0010] In one possible embodiment, the preset control function includes: ; in, For the current control moment, Let water temperature be the objective function. For the prediction cycle, To predict the step size, For the first Water temperature adjustment weights for each forecast period For the first Water temperature output value for each prediction cycle, Let the objective function be the fan speed. For the first Fan speed adjustment weights for each prediction period For the first Fan speed control quantity for each prediction cycle.
[0011] In one possible embodiment, the preset control function includes: ; in, To constrain the unsafe relaxed objective function, As slack variables, For penalty weights.
[0012] Through slack variables With penalty weight The synergistic effect of these factors enables a flexible constraint optimization strategy, resolving the solution dilemma caused by overly strict hard constraints. This is achieved by incorporating relaxation variables into the preset control function. Related penalty items Then, the optimization process will automatically weigh the "cost of violating constraints" against the "cost of decreased control performance," and penalize the terms. The larger the constraint, the higher the cost of violating it outweighs the cost of degrading control performance, and the more the algorithm tends to choose... Smaller solutions (closer to the original constraints) are combined to allow for finite violations to break the solution deadlock, while quantification and penalty mechanisms prevent unlimited deviations.
[0013] In one possible embodiment, obtaining the predicted coolant temperature of the vehicle engine based on the predicted vehicle data includes: The predicted water temperature is obtained based on the predicted vehicle data using a water temperature prediction model, which is either a transfer function model or a linear state-space model.
[0014] Secondly, this application provides a fan control device for a vehicle engine, comprising: A vehicle data acquisition module, which is used to acquire historical vehicle data within a set historical time period; A water temperature prediction module is used to predict the vehicle data for a set future time period based on the historical vehicle data, and to obtain the predicted water temperature of the vehicle engine based on the predicted vehicle data. A fan control module is used to solve a preset control function based on the predicted coolant temperature to obtain the target speed of the fan, and to control the fan based on the target speed. The preset control function is a target function of the vehicle's engine coolant temperature and fan speed.
[0015] Thirdly, this application provides an engine, comprising: An engine body and a cooling system connected to the engine body, the cooling system including a water-cooled structure, a fan, and a fan control device for a vehicle engine as described in claim 8, the fan control device for the vehicle engine being used to control the fan by performing the aforementioned fan control method for a vehicle engine.
[0016] Fourthly, this application provides an automobile, including: the engine as described above.
[0017] The beneficial effects of this application are: Compared with related technologies, the vehicle engine fan control method, device, engine, and automobile provided in this application collect vehicle data as historical vehicle data during vehicle operation. When controlling the fan, historical vehicle data within a set historical time period is obtained. Based on the historical vehicle data, predicted vehicle data for a set future time period is predicted. Based on the predicted vehicle data, the engine coolant temperature is predicted to obtain the future predicted coolant temperature. The target speed of the fan is determined based on the predicted coolant temperature, and the fan is controlled based on the target speed. This enables advance control of the fan, reduces the response time of the fan control process to changes in operating conditions, and solves the technical problems of significantly delayed fan control response and high power consumption. Attached Figure Description
[0018] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0019] Figure 1 This is a schematic flowchart of a vehicle engine fan control method provided in one embodiment of this application; Figure 2 , Figure 3 This is a schematic diagram illustrating the change of water temperature and fan power over time in a vehicle engine fan control method provided in one embodiment of this application. Figure 4 This is a schematic flowchart of a vehicle engine fan control method provided in another embodiment of this application; Figure 5 This is a schematic diagram of the structure of an engine provided in one embodiment of this application; Figure 6 This is a schematic diagram of the structure of a car provided in one embodiment of this application. Detailed Implementation
[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.
[0021] In the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more. "And / or" describes the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone.
[0022] The terms "first," "second," etc., used in the embodiments of this application are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a technical feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature.
[0023] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.
[0024] This application provides a fan control method, device, engine, and automobile for a vehicle engine, which will be described in detail below.
[0025] Please refer to Figure 1 , Figure 2 , Figure 3 The vehicle engine fan control method provided in this embodiment may specifically include the following steps: Step S101: Obtain historical vehicle data for the vehicle within a set historical time period.
[0026] In this embodiment, during vehicle operation, real-time vehicle operation data is collected and stored using the collection time as an identifier. When historical vehicle data is needed, pre-collected and stored historical vehicle data within a set historical time period is retrieved according to actual requirements.
[0027] Specifically, the historical time period can be selected based on the set duration. For example, if 10 minutes of historical vehicle data is needed, the current time can be used as the starting point to obtain vehicle data within the last 10 minutes as historical vehicle data; if 5 minutes of historical vehicle data is needed, the current time can be used as the starting point to obtain vehicle data within the last 5 minutes as historical vehicle data, and so on.
[0028] Specifically, in this embodiment, historical vehicle data may include a series of data such as vehicle speed, engine speed, engine torque, ambient temperature, fan speed, and water temperature.
[0029] Step S102: Predict vehicle data for a set future time period based on historical vehicle data.
[0030] In this embodiment, the predicted vehicle data can be obtained by searching for the mapped vehicle data corresponding to historical vehicle data from a preset correspondence relationship. The preset correspondence relationship is the correspondence between vehicle data obtained from engine universal testing and other vehicle data. Specifically, universal testing is performed on the same type and / or the same vehicle and / or engine in advance. Based on the test data, a correspondence relationship is constructed between vehicle operation data for a previous period and vehicle operation data for a subsequent period. Then, the historical vehicle data obtained in step S101 is substituted into the preset correspondence relationship to search for the data corresponding to the historical vehicle data, which is then used as the predicted vehicle data.
[0031] By pre-determining the correspondence between data from different vehicles through universal engine testing, the correspondence between different vehicle data can be made more realistic, thereby making the predicted vehicle data more realistic and improving the accuracy of the final fan control.
[0032] Furthermore, in this embodiment, historical vehicle data and predicted vehicle data can be of the same type, such as both being engine speed, both being vehicle speed, both being engine heat dissipation, etc.
[0033] In some other embodiments of this application, historical vehicle data and predicted vehicle data can also be different types of vehicle data. For example, historical vehicle data may include engine torque, engine speed, etc., while predicted vehicle data may include engine heat dissipation, vehicle speed, etc. The specific settings can be flexibly configured according to actual needs. Furthermore, the preset correspondence can also be preset based on the preset requirements of historical vehicle data and predicted vehicle data. For example, the preset correspondence could be set as the correspondence between engine torque in a previous period and engine heat dissipation in a subsequent period.
[0034] It is understood that the foregoing is merely an example of a specific method for determining predicted vehicle data based on a preset correspondence in one embodiment of this application. In some other embodiments of this application, for example, it may involve obtaining a vehicle data prediction model pre-trained based on sample vehicle data, inputting the historical vehicle data into the vehicle data prediction model, and obtaining the output result of the vehicle data prediction model as the predicted vehicle data. Specifically, the sample vehicle data can be actual operating data of vehicles of the same type. In the process of training the vehicle data prediction model based on the actual operating data, the sample vehicle-to-everything (V2X) data can be divided into a training dataset and a validation dataset. The vehicle data prediction model is trained based on the training dataset, and the training results are validated based on the validation dataset. This process is repeated until the vehicle data prediction model converges, resulting in a pre-trained vehicle data prediction model.
[0035] In this embodiment, the vehicle data prediction model is specifically a neural network model built based on a Long Short-Term Memory (LSTM) network. LSTM is an improved model of Recurrent Neural Networks (RNNs), particularly suitable for predicting time-series data containing long-term dependencies. Its core structure is the "memory unit," which acts like a "conveyor belt" for information, stably transmitting key information across time steps. Three "gating mechanisms" (input gate, forget gate, and output gate) around the memory unit are responsible for information filtering: the forget gate uses the sigmoid function to discard irrelevant information from historical memory; the input gate combines the current input with short-term memory to filter new information and store it in the memory unit; and the output gate generates the current output based on the updated memory unit. In data prediction, LSTM dynamically adjusts information retention through gating, effectively capturing long-term time-series patterns and is better suited to handling complex data compared to traditional time-series models.
[0036] Taking historical vehicle data such as engine speed, engine torque, and vehicle speed as an example, and predicting vehicle data such as engine cooling data as an example.
[0037] Since there is a certain relationship between data such as engine speed, engine torque, and vehicle speed and engine heat dissipation data, when determining the predicted vehicle data, it is first necessary to perform multi-source data fusion on data such as engine speed, engine torque, and vehicle speed.
[0038] The goal of multi-source data fusion is not to "calculate" engine cooling data from data such as engine speed, engine torque, and vehicle speed using a certain formula, but to build a high-quality, structured input dataset for the LSTM model. The core of this process is time alignment and feature construction, enabling the model to discover the complex relationships between these variables and the cooling data.
[0039] First, create the main timeline: define a target sampling frequency (e.g., 10Hz, 20Hz, etc.). The target sampling frequency is usually an intermediate frequency chosen based on the data frequencies of all data. Frequency alignment is then performed on data such as engine speed, engine torque, and vehicle speed at different frequencies based on the target sampling frequency. For example, for high-frequency signals, the last valid value is taken in each sampling period, while for low-frequency signals, forward padding is used. That is, once a new signal value arrives, this value will continue to pad until all time points before the next new value arrives.
[0040] Next, feature standardization is performed. Because data such as engine speed, engine torque, and vehicle speed differ greatly in units and numerical range (e.g., engine speed is in the thousands of revolutions per minute, while vehicle speed is in the tens of revolutions per minute), directly inputting them into the model would cause the features with large numerical values to dominate the training process, leading to model bias. Therefore, each feature must be standardized, converting it into a distribution with a mean of 0 and a standard deviation of 1.
[0041] Once the multi-source data fusion is complete, that is, after the model input data is constructed, the corresponding processing of the LSTM model can be performed.
[0042] Since LSTM models do not directly process long time series, but instead learn local patterns through a "sliding window" approach, the time step (look-back window) needs to be defined before using an LSTM model. The time step specifies how much historical data the LSTM model needs to observe to make predictions. For example, choosing 60 time steps (if the data is 1 second per record, i.e., the past minute) means that the LSTM model will consider all feature changes within the previous 60 seconds when making predictions. This window length needs to be adjusted according to the dynamic characteristics of the system; too short a window will not see enough historical information, while too long a window will include redundant information and increase the computational burden.
[0043] After setting up the LSTM model, the engine speed, engine torque, vehicle speed, and other data from multi-source data fusion are input into the LSTM model. The LSTM model first uses the input data as samples and constructs a correspondence between samples (X) and labels (y) based on the input data. Taking the time step of 60 as an example, each sample X_i is constructed as a two-dimensional array (or matrix) with the shape (60, n_features). It represents the values of all feature variables within 60 seconds up to a certain time point t. The label y_i corresponding to this sample is the future value to be predicted. If it is a single-step prediction, y_i is the target value (temperature) at time t+1; if it is a multi-step prediction, y_i is a vector containing the target values at multiple future time points from t+1 to t+T. By sliding this window, tens of thousands of such sample-label pairs can be generated to complete the prediction process for vehicle data.
[0044] Alternatively, in some embodiments of this application, the two methods described above can be combined. The preset correspondence can be the correspondence between different vehicle data at the same time, such as the correspondence between engine speed, engine torque, vehicle speed and engine cooling requirements. After obtaining the mapped vehicle data based on historical vehicle data and the preset correspondence, the mapped vehicle data is fitted with data. The data fitting function is extended to a preset time period to obtain the predicted vehicle data corresponding to the preset time period.
[0045] Step S103: Obtain the predicted coolant temperature of the vehicle engine based on the predicted vehicle data.
[0046] In this step, a relationship model between vehicle data and engine coolant temperature can be pre-built. For example, the relationship model can be built based on mathematical methods such as transfer functions and linear state-space functions. After determining the predicted vehicle data in step S102, the predicted vehicle data can be input into the pre-built relationship model to obtain the predicted coolant temperature of the vehicle engine output by the model.
[0047] Alternatively, in some other embodiments of this application, the predicted water temperature can be determined by constructing a correspondence between vehicle data and engine water temperature based on a universal experiment, and then substituting the predicted vehicle data into this correspondence to obtain the predicted water temperature, or by other methods.
[0048] Step S104: Solve the preset control function based on the predicted water temperature to obtain the target speed of the fan, and control the fan based on the target speed. The preset control function is the target function of the vehicle's engine water temperature and fan speed.
[0049] In this application, the preset control function can specifically be: ;in, For the current control moment, Let water temperature be the objective function. For the prediction cycle, To predict the step size, For the first Water temperature adjustment weights for each forecast period For the first Water temperature output value for each prediction cycle, Let fan speed be the objective function. For the first Fan speed adjustment weights for each prediction period No. Fan speed control quantity for each prediction cycle.
[0050] For the aforementioned preset control function, the solution to the preset control function based on the predicted water temperature can be specifically achieved by substituting the predicted water temperature into the preset control function and finding the solution that makes the preset control function reach its minimum value. The corresponding fan speed is then taken as the target fan speed.
[0051] Furthermore, based on the aforementioned preset control function, when it is necessary to strengthen the control weight of the fan speed, the [weight] is increased. When it is necessary to increase the weight of output water temperature control, increase This means that the weight values of different sub-items of the preset control function can be adjusted according to different control requirements.
[0052] Once the target fan speed is determined, control operations such as frequency adjustment and power adjustment can be performed on the fan to ensure that the fan reaches the target speed within a preset future time period, thus achieving advance control of the fan.
[0053] It is understood that the above is merely a specific example of the preset control function provided in this embodiment and does not constitute a limitation. In some embodiments of this application, the preset control function may also be, for example, ;in, To constrain the unsafe relaxed objective function, As slack variables, For penalty weights.
[0054] Among them, relaxation constraints are achieved through slack variables. With penalty weight The synergistic effect of these factors enables a flexible optimization strategy, resolving the problem-solving dilemma caused by overly strict hard constraints. Slack variables are used to transform "strictly satisfied" hard constraints into quantifiable soft constraints. By introducing slack constraints, the originally unsolvable rigid constraint problem is transformed into a feasible problem with measurable "deviation degree." Penalty weights are used to control the reasonableness of the "deviation degree." These slack variables are then added to the preset control function. Related penalty items Then, the optimization process will automatically weigh the "cost of violating constraints" against the "cost of decreased control performance," and penalize the terms. The larger the constraint, the higher the cost of violating it outweighs the cost of degrading control performance, and the more the algorithm tends to choose... Smaller solution (closer to the original constraint); penalty term It can be adjusted according to actual needs, balancing feasibility and constraint fit. The combination of the two allows for limited violations to break through the deadlock, while preventing unlimited deviations through quantification and penalty mechanisms.
[0055] Because there are significant differences in the operating conditions, actual operating status, and actual operating environment of different vehicles, relaxation constraints are introduced into the preset control function. This makes it easier to solve the preset control function, reducing the computing power and time required for solving the preset control function. The reduced computing power requirement makes the fan control method for vehicle engines provided in this application more widely applicable.
[0056] Compared with related technologies, the vehicle engine fan control method, device, engine, and automobile provided in this embodiment collect vehicle data as historical vehicle data during vehicle operation. When controlling the fan, historical vehicle data within a set historical time period is obtained. Based on the historical vehicle data, predicted vehicle data for a set future time period is predicted. Based on the predicted heat dissipation data and predicted driving data, the engine coolant temperature is predicted to obtain the future predicted coolant temperature. The target speed of the fan is determined based on the predicted coolant temperature, and the fan is controlled based on the target speed. This enables advance control of the fan, reduces the response time of the fan control process to changes in operating conditions, and solves the technical problems of significantly delayed fan control response and high power consumption.
[0057] To better implement the vehicle engine fan control method in the embodiments of this application, based on the vehicle engine fan control method, correspondingly, as follows: Figure 4 As shown in the illustration, this application also provides a fan control device for a vehicle engine, the fan control device for the vehicle engine including: The vehicle data acquisition module 301 is used to acquire historical vehicle data within a set historical time period; the water temperature prediction module 302 is used to predict the vehicle data for a set future time period based on the historical vehicle data, and obtain the predicted water temperature of the vehicle engine based on the predicted vehicle data; the fan control module 303 is used to solve a preset control function based on the predicted water temperature to obtain the target speed of the fan, and control the fan based on the target speed. The preset control function is the target function of the vehicle's engine water temperature and fan speed.
[0058] Specifically, the water temperature prediction module 302 includes a vehicle data prediction submodule 3021 for predicting vehicle data for a set future time period based on historical vehicle data, and a water temperature prediction submodule 3022 for obtaining the predicted water temperature of the vehicle engine based on the predicted vehicle data.
[0059] The vehicle engine fan control device provided in the above embodiments can realize the technical solutions described in the above vehicle engine fan control method embodiments. The specific implementation principles of each module or unit can be found in the corresponding content in the above vehicle engine fan control method embodiments, and will not be repeated here.
[0060] Compared with related technologies, the vehicle engine fan control device provided in this embodiment uses a vehicle data acquisition module 301 to collect vehicle data as historical vehicle data. When the fan control module 303 controls the fan, the water temperature prediction module 302 acquires the historical vehicle data collected by the vehicle data acquisition module 301 within a set historical time period. The water temperature prediction module 302 predicts the vehicle data for a set future time period after a certain period based on the historical vehicle data, and predicts the water temperature of the vehicle engine based on the predicted vehicle data to obtain the future predicted water temperature. The fan control module 303 determines the target speed of the fan based on the predicted water temperature and controls the fan based on the target speed. This enables advance control of the fan, reduces the response time of the fan control process to changes in operating conditions, and solves the technical problems of significantly delayed fan control response and high power consumption.
[0061] For further details, please refer to Figure 5This application also provides an engine, including: an engine body 401 and a cooling system 402 connected to the engine body 401. The cooling system 402 includes a water-cooled structure 403 (including a cooling water passage 4021, a water pump 4022, a thermostat 4023, and a radiator 4024) and a fan 404, as well as a fan control device for a vehicle engine as provided in the foregoing embodiments (not shown in the figure). The fan control device for the vehicle engine is used to execute the fan control method for a vehicle engine as provided in the foregoing embodiments to control the fan 404.
[0062] Specifically, in this embodiment, the engine body can be a traditional fuel engine, such as a gasoline engine used in passenger cars, a diesel engine used in trucks, or an electric engine used in hybrid or electric vehicles, as long as it includes a water-cooling structure 403 and a fan 404.
[0063] Furthermore, embodiments of this application also provide an automobile, including: an engine as provided in the foregoing embodiments.
[0064] The above provides a detailed description of the vehicle engine fan control method, device, engine, and automobile provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.
Claims
1. A fan control method for a vehicle engine, characterized in that, include: Retrieve historical vehicle data within a specified historical time period; Based on the historical vehicle data, predict the vehicle data for a set future time period; The predicted engine coolant temperature is obtained based on the predicted vehicle data. The preset control function is solved based on the predicted water temperature to obtain the target speed of the fan. The fan is then controlled based on the target speed. The preset control function is the target function of the vehicle's engine water temperature and fan speed.
2. The fan control method for a vehicle engine according to claim 1, characterized in that, The step of predicting the vehicle data for a set future time period based on the historical vehicle data includes: The predicted vehicle data is obtained based on a preset correspondence and the historical vehicle data. The preset correspondence is the correspondence between different vehicle data obtained from the engine universal test.
3. The vehicle engine fan control method according to claim 1, characterized in that, The step of predicting the vehicle data for a set future time period based on the historical vehicle data includes: Obtain a vehicle data prediction model pre-trained based on sample vehicle data, input the historical vehicle data into the vehicle data prediction model, and obtain the output of the vehicle data prediction model as the predicted vehicle data.
4. The fan control method for a vehicle engine according to claim 3, characterized in that, The vehicle data prediction model is a neural network model built on a long short-term memory network.
5. The fan control method for a vehicle engine according to claim 1, characterized in that, The preset control function includes: ； in, For the current control moment, Let water temperature be the objective function. For the prediction cycle, To predict the step size, For the first Water temperature adjustment weights for each forecast period For the first Water temperature output value for each prediction cycle, Let the objective function be the fan speed. For the first Fan speed adjustment weights for each prediction period For the first Fan speed control quantity for each prediction cycle.
6. The fan control method for a vehicle engine according to claim 5, characterized in that, The preset control function includes: ； in, To constrain the unsafe relaxed objective function, As slack variables, For penalty weights.
7. The fan control method for a vehicle engine according to claim 1, characterized in that, The step of obtaining the predicted coolant temperature of the vehicle engine based on the predicted vehicle data includes: The predicted water temperature is obtained based on the predicted vehicle data using a water temperature prediction model, which is either a transfer function model or a linear state-space model.
8. A fan control device for a vehicle engine, characterized in that, include: A vehicle data acquisition module, which is used to acquire historical vehicle data within a set historical time period; A water temperature prediction module is used to predict the vehicle data for a set future time period based on the historical vehicle data, and to obtain the predicted water temperature of the vehicle engine based on the predicted vehicle data. A fan control module is used to solve a preset control function based on the predicted coolant temperature to obtain the target speed of the fan, and to control the fan based on the target speed. The preset control function is a target function of the vehicle's engine coolant temperature and fan speed.
9. An engine, characterized in that, include: An engine body and a cooling system connected to the engine body, the cooling system including a water-cooled structure, a fan, and a fan control device for a vehicle engine as described in claim 8, the fan control device for the vehicle engine being used to control the fan by performing a fan control method for a vehicle engine as described in any one of claims 1 to 7.
10. A car, characterized in that, include: The engine as described in claim 9.