Control device, method, system and vehicle for vehicle refrigerator

Abstract

The embodiment of the present invention provides a control device, method, system and automobile for a vehicle refrigerator, wherein the control device for the vehicle refrigerator comprises: a machine learning model module, which is used to obtain historical data within a historical time period and real-time data at a first moment; and a machine learning model for training the machine learning model module based on the historical data so that the machine learning model converges; and a method for predicting the temperature change in the vehicle refrigerator after the first moment based on the real-time data to obtain a prediction result; and a method for calculating the gas supply amount Qcoo l and supply time tcoo l that need to be adjusted by the flow control module based on the prediction result; and a regulating module, which is used to control the temperature in the vehicle refrigerator by the flow control module to reach the set target temperature Tset in the vehicle refrigerator at a second moment based on the gas supply amount Qcoo l and supply time tcoo l that need to be adjusted. Through the embodiment of the present invention, the problem that the high energy consumption of the vehicle refrigerator used in the related art brings a high burden to the automobile is solved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the automotive field, and in particular, to a control device, method, system and automobile for a vehicle refrigerator. Background Art

[0002] With the development of modern automobile technology, car refrigerators have become one of the standard or optional equipment of many vehicles, providing great convenience for drivers and passengers. Car refrigerators use the vehicle's power system to keep food and drinks fresh during travel, and can even store medicines and other items that need to be refrigerated. From the initial simple cooling box to the modern electronic temperature-controlled refrigerator, car refrigerators have undergone significant technological progress, not only in terms of cooling efficiency, but also in terms of design, which is increasingly in line with the beauty and practicality of vehicle interiors.

[0003] However, the power consumption of car refrigerators is relatively large, which has an impact on the driving performance of the car to a certain extent. Since car refrigerators need to consume the vehicle's electrical energy to maintain their cooling function, when the refrigerator's energy consumption is too high, it may cause a heavier burden on the vehicle's battery, thereby affecting the vehicle's starting speed and overall power supply. In extreme cases, high-energy-consuming car refrigerators may even cause instability in the vehicle's electrical system, increasing the risk of failure. In addition, in order to meet the power needs of the car refrigerator, the vehicle may need to be equipped with a larger capacity battery or a more efficient generator, which not only increases the cost of the vehicle, but also increases the weight of the vehicle, thereby affecting fuel efficiency, the electric power supply time of the electric vehicle, and environmental friendliness.

[0004] Therefore, although car refrigerators bring convenience to drivers and passengers, their high energy consumption problem needs to be solved urgently. Summary of the invention

[0005] The embodiments of the present invention provide a control device, method, system and vehicle for a vehicle refrigerator, so as to at least solve the problem that the high energy consumption of the vehicle refrigerator used in the related art brings a high burden to the vehicle.

[0006] According to one embodiment of the present invention, a control device for a vehicle refrigerator is provided, comprising:

[0007] A machine learning model module, used to obtain historical data within a historical time period and real-time data at a first moment, wherein the historical data and the real-time data both include: an external environment temperature Text of the vehicle refrigerator, a real-time temperature Tbox inside the vehicle refrigerator, a set target temperature Tset inside the vehicle refrigerator, a vehicle speed Vveh, and an ideal storage temperature Titem of items in the vehicle refrigerator; and,

[0008] for training the machine learning model of the machine learning model module based on the historical data so that the machine learning model converges; and,

[0009] used to predict the temperature change in the vehicle refrigerator after the first moment based on the real-time data to obtain a prediction result; and

[0010] Used to calculate the gas supply quantity Qcool and supply time tcool that need to be adjusted by the flow control module based on the prediction result, so as to control the flow of gas provided by the vehicle air conditioning system to the vehicle refrigerator;

[0011] a regulating module, configured to enable the flow control module to control the temperature in the vehicle refrigerator to reach the set target temperature Tset in the vehicle refrigerator at a second moment based on the gas supply quantity Qcool and the supply time tcool that need to be adjusted;

[0012] Wherein, the temperature prediction algorithm for calculating the prediction result is: Tpred=f(Text, Vveh, Tbox, Titem), wherein Tpred is the prediction result, and f is the prediction function of the machine learning model;

[0013] The algorithm for calculating the gas supply amount Qcool is: Qcool=α*|Tpred-Tbox|, where α is an adjustment factor for optimizing the gas flow rate according to the specific conditions and historical data of the vehicle;

[0014] The algorithm for calculating the supply time tcool is: tcool=β*Qcool, where β is a time adjustment factor used to determine the time required to reach the set target temperature Tset in the vehicle refrigerator;

[0015] The control device is connected to the vehicle air-conditioning system and the vehicle refrigerator, and the flow control module is connected to the control device, the vehicle air-conditioning system and the vehicle refrigerator.

[0016] According to another embodiment of the present invention, there is also provided a control method for a vehicle refrigerator, comprising:

[0017] Acquire historical data within a historical time period to train the machine learning model of the machine learning model module based on the historical data so as to converge the machine learning model;

[0018] Acquire real-time data at a first moment, and predict the temperature change in the vehicle refrigerator after the first moment based on the real-time data to obtain a prediction result;

[0019] Based on the prediction result, the gas supply quantity Qcool and the supply time tcool that the flow control module needs to adjust are calculated to control the flow of gas provided by the vehicle air conditioning system to the vehicle refrigerator;

[0020] Based on the gas supply quantity Qcool and the supply time tcool that need to be adjusted, the flow control module controls the temperature in the vehicle refrigerator to reach the set target temperature Tset in the vehicle refrigerator at the second moment;

[0021] Wherein, the temperature prediction algorithm for calculating the prediction result is: Tpred=f(Text, Vveh, Tbox, Titem), wherein Tpred is the prediction result, and f is the prediction function of the machine learning model;

[0022] The algorithm for calculating the gas supply amount Qcool is: Qcool=α*|Tpred-Tbox|, where α is an adjustment factor for optimizing the gas flow rate according to the specific conditions and historical data of the vehicle;

[0023] The algorithm for calculating the supply time tcool is: tcool=β*Qcool, where β is a time adjustment factor used to determine the time required to reach the set target temperature Tset in the vehicle refrigerator;

[0024] The control device is connected to the vehicle air-conditioning system and the vehicle refrigerator, and the flow control module is connected to the control device, the vehicle air-conditioning system and the vehicle refrigerator.

[0025] According to another embodiment of the present invention, a control system for a vehicle refrigerator is provided, comprising:

[0026] A control device as any of the above;

[0027] A vehicle air conditioning system for generating gas, wherein the gas includes cold gas or hot gas;

[0028] a flow control module, used for adjusting the flow of the gas based on the control device;

[0029] A vehicle refrigerator is used to change the temperature inside the vehicle refrigerator based on the conditioned gas.

[0030] According to yet another embodiment of the present invention, there is provided a car comprising any control system as described above.

[0031] Through one of the embodiments of the present invention, firstly, the machine learning model module adopted can learn and predict according to historical data and real-time data (including external environment temperature, real-time temperature in the refrigerator, target temperature, vehicle driving speed and ideal storage temperature of items). This intelligent prediction method can more accurately predict the temperature change in the refrigerator, thereby reducing unnecessary energy consumption caused by temperature fluctuations. Secondly, according to the prediction results, the present invention can dynamically adjust the gas supply amount and supply time to accurately control the flow rate of cooling gas provided by the vehicle air conditioning system to the refrigerator. This precise control can avoid overcooling or insufficient cooling, thereby reducing energy consumption. Furthermore, the present invention adopts a specific algorithm to calculate the gas supply amount and supply time, which includes adjustment factors and time adjustment factors, which can be optimized according to the specific conditions and historical data of the vehicle to further improve energy efficiency. Furthermore, the control device of the present invention is directly connected to the vehicle air conditioning system and the vehicle refrigerator, realizing seamless integration between systems. This integration allows more efficient energy utilization, because the vehicle air conditioning system can adjust its working state according to the real-time needs of the refrigerator, rather than continuously operating in a fixed mode. Furthermore, through the above-mentioned intelligent control and optimization algorithm, the present invention can significantly reduce the burden of the vehicle refrigerator on the vehicle power system. This not only reduces the pressure on the battery, but also improves the overall power efficiency of the vehicle, avoiding the problems of reduced startup speed and unstable power supply caused by the high energy consumption of the refrigerator. Furthermore, since the present invention reduces the need for larger capacity batteries or more efficient generators, it helps to reduce the cost of the vehicle and reduce the impact on the environment, because more efficient energy use reduces dependence on fossil fuels or the car's generator. In summary, the present invention effectively solves the problem of high energy consumption of car refrigerators through intelligent temperature prediction and dynamic flow control, combined with optimization algorithms and system integration, improves the overall energy efficiency and performance of the car, and also complies with the modern car design concept of environmental protection and energy saving. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 Schematic diagram of the structure of the control device of the vehicle refrigerator according to an embodiment of the present invention Figure 1 ;

[0033] Figure 2 is a schematic diagram of the connection relationship among a control device for a vehicle refrigerator, a flow control module, a vehicle air conditioning system, and a vehicle refrigerator according to an embodiment of the present invention;

[0034] Figure 3 Schematic diagram of the structure of the control device of the vehicle refrigerator according to an embodiment of the present invention Figure 2 ;

[0035] Figure 4 is a flow chart of a control method of a vehicle refrigerator according to an embodiment of the present invention.

[0036] Explanation of the accompanying drawings: 1. Control device of vehicle refrigerator; 11. Machine learning model module; 12. Adjustment module; 13. Closed-loop control module; 2. Flow control module; 3. Vehicle air-conditioning system; 4. Vehicle refrigerator. DETAILED DESCRIPTION

[0037] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings and in combination with the embodiments.

[0038] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0039] In this embodiment, a control device for a vehicle refrigerator is provided. Figure 1 Schematic diagram of the structure of the control device of the vehicle refrigerator according to an embodiment of the present invention Figure 1 , Figure 2 Schematic diagram of the connection relationship among the control device of the vehicle refrigerator, the flow control module, the vehicle air conditioning system, and the vehicle refrigerator according to an embodiment of the present invention. Figure 1 and Figure 2 As shown, the control device 1 of the vehicle refrigerator includes:

[0040] The machine learning model module 11 is used to obtain historical data within a historical time period and real-time data at a first moment, wherein both the historical data and the real-time data include: an external environment temperature Text of the vehicle refrigerator, a real-time temperature Tbox inside the vehicle refrigerator, a set target temperature Tset inside the vehicle refrigerator, a vehicle speed Vveh, and an ideal storage temperature Titem of items in the vehicle refrigerator; and,

[0041] A machine learning model for training the machine learning model module 11 based on historical data to make the machine learning model converge; and

[0042] In an exemplary embodiment, in order to train the machine learning model module 11, the following specific process can be adopted, with the external environment temperature Text of the vehicle refrigerator, the real-time temperature Tbox in the vehicle refrigerator, the set target temperature Tset in the vehicle refrigerator, the vehicle speed Vveh, and the ideal storage temperature Titem of the items in the vehicle refrigerator as feature variables:

[0043] 1. Data collection and processing.

[0044] First, data of the vehicle refrigerator 4 and the vehicle within a certain time range are collected, for example: the external environment temperature Text of the vehicle refrigerator, the real-time temperature Tbox inside the vehicle refrigerator, the set target temperature Tset inside the vehicle refrigerator, the vehicle speed Vveh, and the ideal storage temperature Titem of the items in the vehicle refrigerator.

[0045] Clean the data, including removing outliers, filling missing values, and standardizing the data.

[0046] 2. Feature engineering.

[0047] Feature selection: Select features that have a significant impact on temperature prediction. This can include all the above variables.

[0048] Feature construction: New features can be constructed, such as temperature difference: |Tset-Tbox|, interaction terms between speed and temperature difference, etc.

[0049] 3. Model selection.

[0050] Choose a machine learning model suitable for time series forecasting, such as a long short-term memory network (LSTM):

[0051] LSTM model: suitable for processing and predicting time series data, and can capture long-term dependencies in time series.

[0052] 4. Model training.

[0053] Data partitioning: Divide the data into training and testing sets, for example, 75% of the data is used for training and 25% for testing.

[0054] Model construction: Build an LSTM model, where the input layer receives feature variables and the output layer outputs the predicted temperature.

[0055] Hyperparameter tuning: Adjust the hyperparameters of LSTM, such as learning rate, number of neurons in the hidden layer, number of training iterations, etc., to optimize model performance.

[0056] 5. Model Evaluation

[0057] Cross-validation: Use cross-validation method to evaluate the stability and accuracy of the model.

[0058] Performance indicators: Use indicators such as mean square error (MSE) and root mean square error (RMSE) to evaluate the prediction performance of the model.

[0059] 6. Model optimization.

[0060] Based on the results of the model evaluation, the model structure or hyperparameters are further adjusted to improve the accuracy of the prediction.

[0061] 7. Model deployment.

[0062] The trained model is deployed to the control system of the vehicle refrigerator 4 to receive new data in real time and perform temperature prediction.

[0063] for predicting the temperature change in the vehicle refrigerator 4 after the first moment based on the real-time data to obtain a prediction result; and

[0064] Used to calculate the gas supply quantity Qcool and supply time tcool that the flow control module 2 needs to adjust based on the prediction result, so as to control the flow of gas provided by the vehicle air conditioning system 3 to the vehicle refrigerator 4;

[0065] In an exemplary implementation, after the model deployment process is completed, for example, the following method is adopted:

[0066] 8. Real-time prediction and output.

[0067] Real-time data input: Real-time data is input into the machine learning model module 11 to obtain temperature predictions.

[0068] Output decision: According to the prediction results, the gas supply volume Qcool and supply time tcool that need to be adjusted are output. For example, the prediction results include the mapping relationship between time and gas supply volume Qcool and supply time tcool. The first moment is used as the time starting point, and the gas supply volume Qcool and supply time tcool at other moments are displayed with the preset time period as the time interval. Of course, it can also be the gas supply volume Qcool and supply time tcool in the time period from the first moment to the second moment. That is: display the gas supply volume Qcool and supply time tcool at a certain moment, or display the gas supply volume Qcool and supply time tcool in a certain time period.

[0069] Through the above process, the machine learning model is able to learn the pattern of temperature change of the vehicle refrigerator 4 under different conditions and is able to predict future temperature changes, thereby achieving more accurate and energy-saving temperature control.

[0070] For example, the predicted temperature of the prediction result is compared with the target temperature Tset set in the car refrigerator in real time. When the predicted temperature of the prediction result is equal to the target temperature Tset set in the car refrigerator, the time / time period, gas supply volume Qcool and supply time tcool corresponding to the predicted temperature of the prediction result are output.

[0071] Alternatively, the machine learning model module 11 takes the target temperature Tset set in the car refrigerator as the first input at the first moment, automatically collects the external environment temperature Text of the car refrigerator at the first moment, the real-time temperature Tbox in the car refrigerator, the vehicle speed Vveh, and the ideal storage temperature Titem of the items in the car refrigerator as the second input, and outputs the corresponding time / time period, gas supply volume Qcool, and supply time tcool based on the model algorithm output when the predicted temperature of the prediction result is equal to the target temperature Tset in the car refrigerator set at the first moment.

[0072] The regulating module 12 is used to control the temperature in the vehicle refrigerator 4 to reach the set target temperature Tset in the vehicle refrigerator at the second moment based on the gas supply quantity Qcool and the supply time tcool that need to be adjusted by the flow control module 2;

[0073] In an exemplary embodiment, the adjustment module 12 controls the flow control module 2 to perform corresponding operations based on the time / time period, gas supply volume Qcool, and supply time tcool output by the machine learning model module 11 to change the gas flow obtained by the vehicle refrigerator 4.

[0074] The temperature prediction algorithm for calculating the prediction result is: Tpred = f(Text, Vveh, Tbox, Titem), where Tpred is the prediction result and f is the prediction function of the machine learning model;

[0075] The algorithm for calculating the gas supply Qcool is: Qcool = α*|Tpred-Tbox|, where α is an adjustment factor used to optimize the gas flow rate according to the specific conditions and historical data of the vehicle;

[0076] The algorithm for calculating the supply time tcool is: tcool = β*Qcool, where β is a time adjustment factor used to determine the time required to reach the set target temperature Tset in the vehicle refrigerator;

[0077] The control device is connected to the vehicle air conditioning system 3 and the vehicle refrigerator 4 , and the flow control module 2 is connected to the control device, the vehicle air conditioning system 3 and the vehicle refrigerator 4 .

[0078] For example, Figure 2 As shown, the control device 1 of the vehicle refrigerator is connected to the vehicle air-conditioning system 3 and the vehicle refrigerator 4 , and the flow control module 2 is connected to the control device 1 of the vehicle refrigerator, the vehicle air-conditioning system 3 and the vehicle refrigerator 4 .

[0079] Through the embodiments of the present invention, firstly, the machine learning model module 11 can learn and predict based on historical data and real-time data (including external environment temperature, real-time temperature in the refrigerator, target temperature, vehicle speed and ideal storage temperature of items). This intelligent prediction method can more accurately predict the temperature change in the refrigerator, thereby reducing unnecessary energy consumption caused by temperature fluctuations.

[0080] Secondly, according to the prediction results, the present invention can dynamically adjust the gas supply amount and supply time to accurately control the flow rate of cooling gas provided by the vehicle air conditioning system 3 to the refrigerator. This precise control can avoid overcooling or insufficient cooling, thereby reducing energy consumption.

[0081] Furthermore, the present invention uses a specific algorithm to calculate the gas supply amount and supply time, which includes an adjustment factor and a time adjustment factor. These factors can be optimized according to the specific conditions and historical data of the vehicle to further improve energy efficiency.

[0082] Furthermore, the control device of the present invention is directly connected to the vehicle air conditioning system 3 and the vehicle refrigerator 4, achieving seamless integration between the systems. This integration allows for more efficient energy utilization, because the vehicle air conditioning system 3 can adjust its working state according to the real-time needs of the refrigerator, rather than continuously operating in a fixed mode.

[0083] Furthermore, through the above-mentioned intelligent control and optimization algorithm, the present invention can significantly reduce the burden of the vehicle power system caused by the vehicle refrigerator 4. This not only reduces the pressure on the battery, but also improves the overall power efficiency of the vehicle, avoiding the problems of reduced startup speed and unstable power supply caused by the high energy consumption of the refrigerator.

[0084] Furthermore, because the present invention reduces the need for larger batteries or more efficient generators, it helps reduce the cost of the vehicle and reduces the impact on the environment because more efficient energy use reduces reliance on fossil fuels or the vehicle's generator.

[0085] In summary, the present invention effectively solves the problem of high energy consumption of the vehicle refrigerator 4 through intelligent temperature prediction and dynamic flow control, combined with optimization algorithm and system integration, improves the overall energy efficiency and performance of the vehicle, and also complies with the modern automobile design concept of environmental protection and energy saving.

[0086] In one embodiment, the adjustment module 12 is further configured to:

[0087] Obtain the real-time temperature Tbox in the vehicle refrigerator at the third moment, and dynamically adjust the gas supply amount Qcool and the supply time tcool based on the real-time temperature Tbox in the vehicle refrigerator at the third moment, so that the flow control module 2 controls the temperature in the vehicle refrigerator 4 to reach the set target temperature Tset in the vehicle refrigerator at the second moment;

[0088] The third moment is any moment between the first moment and the second moment.

[0089] In an exemplary embodiment, for example, the prediction result includes a mapping relationship between time and gas supply amount Qcool and supply time tcool, with the first moment as the time starting point, and the gas supply amount Qcool and supply time tcool at other moments are displayed at a preset time interval. Of course, it can also be the gas supply amount Qcool and supply time tcool in the time period from the first moment to the second moment. That is: display the gas supply amount Qcool and supply time tcool at a certain moment, or display the gas supply amount Qcool and supply time tcool in a certain time period. This scheme may include the gas supply amount Qcool and supply time tcool in the time period from the first moment to the second moment. At this time, the second moment is the moment when the predicted temperature of the prediction result is equal to the target temperature Tset in the car refrigerator set at the first moment. That is: the gas supply amount Qcool and supply time tcool that need to be adjusted in the time period from the start adjustment moment to the stop adjustment moment are displayed.

[0090] However, there may be a variety of emergencies during vehicle driving. For example, during long-distance driving, extreme weather such as hail, rain, snow, and traffic jams such as sudden traffic accidents may occur. The external environment temperature Text of the vehicle refrigerator, the vehicle speed Vveh, etc. may change rapidly at any time, which may cause the real-time temperature Tbox in the vehicle refrigerator to be difficult to change according to the expected situation of the machine learning model module 11. Then, at any time between the first moment and the second moment (the third moment), the real-time temperature Tbox in the vehicle refrigerator is detected again in real time. If the difference between the real-time temperature Tbox in the vehicle refrigerator at the third moment and the real-time temperature Tbox in the vehicle refrigerator at the third moment expected by the machine learning model module 11 is outside the preset range, then based on the external environment temperature Text of the vehicle refrigerator at the third moment, the real-time temperature Tbox in the vehicle refrigerator, the vehicle speed Vveh, the ideal storage temperature Titem of the items in the vehicle refrigerator, etc., it is predicted again to provide the output time / time period, gas supply amount Qcool, and supply time tcool corresponding to the predicted temperature of the predicted result being equal to the target temperature Tset in the vehicle refrigerator set at the first moment, so as to make real-time adjustments based on the actual situation of the vehicle. Of course, the third moment can be any moment between the first moment and the second moment, or can be multiple moments.

[0091] In one embodiment, the temperature prediction algorithm further includes:

[0092] Tpred=f(Text,Vveh,Tbox,Titem)

[0093] =w1*Text+w2*Vveh+w3*Tbox+w4*Titem+b;

[0094] Among them, w1, w2, w3, and w4 are weight parameters, indicating the influence of each input variable on the prediction result; b is a bias term used to adjust the benchmark of the prediction result.

[0095] In an exemplary embodiment, processing data with different units and dimensions is a common problem in machine learning models. The temperature (Text, Tbox, Tset, Titem) and speed (Vveh) data involved here have different units and dimensions. In order to reasonably set the weight parameters w1, w2, w3, w4 and the bias term b and obtain accurate prediction results Tpred, the following steps can be taken to deal with the problem of inconsistent units:

[0096] 1. Data standardization.

[0097] Data normalization is a common method for dealing with data in different units. It eliminates the effects of different dimensions by scaling the data to a uniform range (usually 0 to 1). Normalization methods include:

[0098] Minimum-maximum normalization: Scale the data to the interval [0,1]. For example, the formula is:

[0099]

[0100] Z-score standardization: convert the mean of the data to 0 and the standard deviation to 1. The formula is:

[0101]

[0102] Among them, μ is the mean and σ is the standard deviation.

[0103] 2. Feature engineering.

[0104] In addition to standardization, feature engineering can also be used to process data in different units:

[0105] Normalization: Dividing data such as speed by a specific constant (such as maximum speed) to reduce the range of values.

[0106] Create new features: For example, you can calculate the temperature difference (Tset-TboxTset-Tbox) or the logarithmic transformation of the velocity, etc.

[0107] 3.Weight adjustment.

[0108] During model training, weights can be adjusted in the following ways:

[0109] Automatic Adjustment: During the training process, most machine learning algorithms (such as linear regression, neural networks) automatically adjust the weights.

[0110] Manual Adjustment: In some cases, the initial value of the weights can be manually set or the adjustment range of the weights can be restricted based on prior knowledge.

[0111] 4. Model validation.

[0112] Use methods such as cross-validation to verify the performance of the model on different datasets to ensure that the model responds appropriately to all features.

[0113] 5. Model tuning.

[0114] Based on the performance of the model on the validation set, adjust the weights and biases. You can use methods such as grid search, random search, or Bayesian optimization to find the optimal parameter combination.

[0115] It should be noted that w1, w2, w3, w4, and b are learned through training data.

[0116] Figure 3 Schematic diagram of the structure of the control device of the vehicle refrigerator according to an embodiment of the present invention Figure 2 In one embodiment, Figure 3 As shown, the control device also includes:

[0117] A closed-loop control module 13 is used to calculate the difference between the adjusted real-time temperature Tbox in the vehicle refrigerator and the set target temperature Tset in the vehicle refrigerator; and

[0118] A control signal is generated based on the difference and the PID controller, so that the flow control module 2 dynamically adjusts the gas supply amount Qcool and the supply time tcool based on the control signal.

[0119] In one embodiment, the control signal generation algorithm includes:

[0120]

[0121] e(t) = |Tpred-Tbox|;

[0122] Among them, e(t) is the difference between the real-time temperature Tbox in the car refrigerator and the predicted result Tpred, Kp is the gain coefficient of e(t), Ki is the integral gain coefficient of e(t), Kd is the differential gain coefficient of e(t), and u(t) is the generated control signal.

[0123] In an exemplary embodiment, the control signal obtained by using the above generation algorithm, that is, an algorithm based on a PID (proportional-integral-differential) controller, has the following advantages and effects:

[0124] Fast response: The proportional (P) part enables the system to respond quickly to deviations, that is, when a temperature deviation (such as e(t)) occurs, the controller immediately generates a control signal (such as u(t)) proportional to the deviation, thereby quickly adjusting the temperature of the vehicle refrigerator 4.

[0125] Reducing overshoot: The integral (I) part adjusts the historical accumulated deviation, which helps to eliminate steady-state errors, ensure that the vehicle refrigerator 4 reaches and maintains the set target temperature for a long time, and reduce overshoot caused by continuous deviation.

[0126] Smoothing control: The derivative (D) part predicts the future trend of the deviation and reduces the fluctuation of temperature change by adjusting the control signal in advance, making the temperature control smoother and reducing the oscillation of the system.

[0127] Strong adaptability: The PID controller can adjust the gain coefficients (Kp, Ki, Kd) according to system characteristics and environmental conditions to adapt to different control requirements and system dynamic changes and improve the robustness of the system.

[0128] Simple and easy to implement: The PID control algorithm is relatively simple, easy to implement and adjust, does not require complex models or a lot of computing resources, and is suitable for embedded systems and real-time control systems.

[0129] Stability and accuracy: By properly adjusting the PID parameters, the stability of the system can be ensured and the accuracy of temperature control can be improved so that the temperature in the vehicle refrigerator 4 is as close to the set target temperature as possible.

[0130] Reduce energy consumption: By precisely controlling the gas supply volume and supply time, unnecessary energy consumption can be avoided, the energy efficiency ratio of the car refrigerator 4 can be improved, the battery life can be extended, and the operating cost can be reduced.

[0131] Real-time adjustment: The closed-loop control module 13 is capable of monitoring and adjusting the temperature of the refrigerator in real time, ensuring that the temperature inside the refrigerator can be quickly adjusted to the target temperature even when the external environment or vehicle driving conditions change.

[0132] Improve user experience: By maintaining a stable temperature inside the refrigerator, the freshness and quality of stored items can be ensured, improving the user experience of drivers and passengers.

[0133] In summary, the control signal generated by the PID control algorithm can achieve fast, accurate and stable temperature control, improve the performance and efficiency of the vehicle refrigerator 4, and reduce energy consumption and improve user experience.

[0134] In one embodiment, the flow control module 2 includes:

[0135] The electronic expansion valve is used to dynamically adjust the gas supply quantity Qcool and supply time tcool based on the control signal.

[0136] In an exemplary embodiment, one end of the electronic expansion valve is connected to the vehicle air-conditioning system 3 through a pipeline, and the other end is connected to the vehicle refrigerator 4 through a pipeline. The electronic expansion valve receives a control signal from the closed-loop control module 13, adjusts the opening and closing degree and opening and closing time, so as to change the flow rate of gas output from the vehicle air-conditioning system 3 to the vehicle refrigerator 4.

[0137] In one embodiment, the electronic multi-way valve includes multiple input ports and output ports, wherein the multiple input ports are connected to the vehicle air-conditioning system 3, and the output ports are connected to the vehicle refrigerator 4, so as to control the opening / closing of the multiple input ports based on a control signal.

[0138] In an exemplary embodiment, the electronic multi-way valve receives a control signal sent by the closed-loop control module 13, and can adjust the opening and closing degree and opening and closing time of multiple input ports. For example, when the vehicle refrigerator 4 needs to change the temperature quickly, the input port closed at the current moment can be changed to an open state to increase the number of open input ports, or increase the opening degree and opening time of the open input port to increase the gas flow received from the vehicle air-conditioning system 3. Similarly, when the vehicle refrigerator 4 does not need to change the temperature quickly, one or more input ports opened at the current moment can be changed to a closed state to reduce the number of open input ports, or reduce the opening degree and opening time of the open input port to reduce the gas flow received from the vehicle air-conditioning system 3. Similarly, the opening degree and opening time of the output port can also be controlled to change the gas flow received by the vehicle refrigerator 4. Of course, the number of output ports can also be one or more, multiple output ports are connected to one vehicle refrigerator 4, or multiple output ports are connected to different vehicle refrigerators 4, which can be set based on actual conditions, and the embodiment of the present invention will not be repeated.

[0139] In one embodiment, the gas passing through the vehicle refrigerator 4 can be transported back to the vehicle air conditioning system 3 or the vehicle's gas circulation system through the flow control module 2 .

[0140] In this embodiment, a control method for a vehicle refrigerator is also provided. Figure 4 is a flow chart of a control method of a vehicle refrigerator according to an embodiment of the present invention. Figure 1 , Figure 2 , Figure 4 As shown, the process includes the following steps:

[0141] Step S401, obtaining historical data within a historical time period to train the machine learning model of the machine learning model module 11 based on the historical data, so as to make the machine learning model converge;

[0142] Step S402, acquiring real-time data at a first moment, and predicting the temperature change in the vehicle refrigerator 4 after the first moment based on the real-time data, so as to obtain a prediction result;

[0143] Step S403, calculating the gas supply amount Qcool and supply time tcool that the flow control module 2 needs to adjust based on the prediction result, so as to control the flow of gas provided by the vehicle air conditioning system 3 to the vehicle refrigerator 4;

[0144] Step S404, based on the gas supply quantity Qcool and the supply time tcool that need to be adjusted, the flow control module 2 controls the temperature in the vehicle refrigerator 4 to reach the set target temperature Tset in the vehicle refrigerator at the second moment;

[0145] The temperature prediction algorithm for calculating the prediction result is: Tpred = f(Text, Vveh, Tbox, Titem), where Tpred is the prediction result and f is the prediction function of the machine learning model;

[0146] The algorithm for calculating the gas supply Qcool is: Qcool = α*|Tpred-Tbox|, where α is an adjustment factor used to optimize the gas flow rate according to the specific conditions and historical data of the vehicle;

[0147] The algorithm for calculating the supply time tcool is: tcool = β*Qcool, where β is a time adjustment factor used to determine the time required to reach the set target temperature Tset in the vehicle refrigerator;

[0148] The control device is connected to the vehicle air conditioning system 3 and the vehicle refrigerator 4 , and the flow control module 2 is connected to the control device, the vehicle air conditioning system 3 and the vehicle refrigerator 4 .

[0149] In an exemplary embodiment, in order to train the machine learning model module 11, the following specific process can be adopted, with the external environment temperature Text of the vehicle refrigerator, the real-time temperature Tbox in the vehicle refrigerator, the set target temperature Tset in the vehicle refrigerator, the vehicle speed Vveh, and the ideal storage temperature Titem of the items in the vehicle refrigerator as feature variables:

[0150] 1. Data collection and processing.

[0151] First, data of the vehicle refrigerator 4 and the vehicle within a certain time range are collected, for example: the external environment temperature Text of the vehicle refrigerator, the real-time temperature Tbox inside the vehicle refrigerator, the set target temperature Tset inside the vehicle refrigerator, the vehicle speed Vveh, and the ideal storage temperature Titem of the items in the vehicle refrigerator.

[0152] Clean the data, including removing outliers, filling missing values, and standardizing the data.

[0153] 2. Feature engineering.

[0154] Feature selection: Select features that have a significant impact on temperature prediction. This can include all the above variables.

[0155] Feature construction: New features can be constructed, such as temperature difference: |Tset-Tbox|, interaction terms between speed and temperature difference, etc.

[0156] 3. Model selection.

[0157] Choose a machine learning model suitable for time series forecasting, such as a long short-term memory network (LSTM):

[0158] LSTM model: suitable for processing and predicting time series data, and can capture long-term dependencies in time series.

[0159] 4. Model training.

[0160] Data partitioning: Divide the data into training and testing sets, for example, 75% of the data is used for training and 25% for testing.

[0161] Model construction: Build an LSTM model, where the input layer receives feature variables and the output layer outputs the predicted temperature.

[0162] Hyperparameter tuning: Adjust the hyperparameters of LSTM, such as learning rate, number of neurons in the hidden layer, number of training iterations, etc., to optimize model performance.

[0163] 5. Model Evaluation

[0164] Cross-validation: Use cross-validation method to evaluate the stability and accuracy of the model.

[0165] Performance indicators: Use indicators such as mean square error (MSE) and root mean square error (RMSE) to evaluate the prediction performance of the model.

[0166] 6. Model optimization.

[0167] Based on the results of the model evaluation, the model structure or hyperparameters are further adjusted to improve the accuracy of the prediction.

[0168] 7. Model deployment.

[0169] The trained model is deployed to the control system of the vehicle refrigerator 4 to receive new data in real time and perform temperature prediction.

[0170] 8. Real-time prediction and output.

[0171] Real-time data input: Real-time data is input into the machine learning model module 11 to obtain temperature predictions.

[0172] Output decision: According to the prediction results, the gas supply volume Qcool and supply time tcool that need to be adjusted are output. For example, the prediction results include the mapping relationship between time and gas supply volume Qcool and supply time tcool. The first moment is used as the time starting point, and the gas supply volume Qcool and supply time tcool at other moments are displayed with the preset time period as the time interval. Of course, it can also be the gas supply volume Qcool and supply time tcool in the time period from the first moment to the second moment. That is: display the gas supply volume Qcool and supply time tcool at a certain moment, or display the gas supply volume Qcool and supply time tcool in a certain time period.

[0173] Through the above method, the machine learning model can learn the pattern of temperature change of the vehicle refrigerator 4 under different conditions and can predict future temperature changes, thereby achieving more accurate and energy-saving temperature control.

[0174] For example, the predicted temperature of the prediction result is compared with the target temperature Tset set in the car refrigerator in real time. When the predicted temperature of the prediction result is equal to the target temperature Tset set in the car refrigerator, the time / time period, gas supply volume Qcool and supply time tcool corresponding to the predicted temperature of the prediction result are output.

[0175] Alternatively, the machine learning model module 11 takes the target temperature Tset set in the car refrigerator as the first input at the first moment, automatically collects the external environment temperature Text of the car refrigerator at the first moment, the real-time temperature Tbox in the car refrigerator, the vehicle speed Vveh, and the ideal storage temperature Titem of the items in the car refrigerator as the second input, and outputs the corresponding time / time period, gas supply volume Qcool, and supply time tcool based on the model algorithm output when the predicted temperature of the prediction result is equal to the target temperature Tset in the car refrigerator set at the first moment.

[0176] Finally, the adjustment module 12 controls the flow control module 2 to perform corresponding operations based on the time / time period, gas supply volume Qcool, and supply time tcool output by the machine learning model module 11 to change the gas flow obtained by the vehicle refrigerator 4.

[0177] Through the above steps S401 to S404, first, through steps S401 and S402, the embodiment of the present invention uses a machine learning model to predict the temperature change of the vehicle refrigerator 4. This prediction is based on historical data and real-time data, including the external environment temperature, the vehicle driving speed, the real-time temperature in the refrigerator, and the ideal storage temperature of the items. The introduction of the machine learning model makes the temperature control more accurate and reduces the additional energy consumption caused by temperature fluctuations.

[0178] In step S403, the embodiment of the present invention dynamically adjusts the gas supply amount and supply time according to the prediction results. This adjustment is based on the algorithm Qcool = α*|Tpred-Tbox| and tcool = β*Qcool, where α and β are adjustment factors for optimizing gas flow and supply time. This dynamic adjustment can reduce unnecessary energy consumption because the gas supply amount and supply time will only be increased when the actual temperature is significantly different from the target temperature.

[0179] The algorithm in the embodiment of the present invention takes into account the specific conditions and historical data of the vehicle, and optimizes the gas flow rate and supply time by adjusting the factor α and the time adjustment factor β. This optimization helps to reduce energy consumption while ensuring that the temperature in the refrigerator can quickly and accurately reach the target temperature.

[0180] The control device of the embodiment of the present invention is directly connected to the vehicle air conditioning system 3 and the vehicle refrigerator 4, and the flow control module 2 is also connected to these systems. This integration allows for more efficient energy utilization because the vehicle air conditioning system 3 can adjust its working state according to the real-time needs of the refrigerator instead of continuously operating in a fixed mode.

[0181] Through the above intelligent control and optimization algorithm, the embodiment of the present invention can significantly reduce the burden of the vehicle power system caused by the vehicle refrigerator 4. This not only reduces the pressure on the battery, but also improves the overall power efficiency of the vehicle, avoiding the problems of reduced startup speed and unstable power supply caused by the high energy consumption of the refrigerator.

[0182] Because embodiments of the present invention reduce the need for larger batteries or more efficient generators, they help reduce the cost of vehicles and lessen the impact on the environment because more efficient energy use reduces dependence on fossil fuels.

[0183] In summary, the embodiment of the present invention effectively solves the problem of high energy consumption of the vehicle refrigerator 4 through intelligent temperature prediction, dynamic flow control, optimization algorithm and system integration, improves the overall energy efficiency and performance of the vehicle, and also complies with the modern automobile design concept of environmental protection and energy saving.

[0184] In one embodiment, the method further comprises:

[0185] Obtain the real-time temperature Tbox in the vehicle refrigerator at the third moment, and dynamically adjust the gas supply amount Qcool and the supply time tcool based on the real-time temperature Tbox in the vehicle refrigerator at the third moment, so that the flow control module 2 controls the temperature in the vehicle refrigerator 4 to reach the set target temperature Tset in the vehicle refrigerator at the second moment;

[0186] The third moment is any moment between the first moment and the second moment.

[0187] In one embodiment, the method further comprises:

[0188] Tpred=f(Text,Vveh,Tbox,Titem)

[0189] =w1*Text+w2*Vveh+w3*Tbox+w4*Titem+b;

[0190] Among them, w1, w2, w3, and w4 are weight parameters, indicating the influence of each input variable on the prediction result; b is a bias term used to adjust the benchmark of the prediction result.

[0191] In one embodiment, the method further comprises:

[0192] Calculate the difference between the adjusted real-time temperature Tbox in the vehicle refrigerator and the set target temperature Tset in the vehicle refrigerator;

[0193] A control signal is generated based on the difference and the PID controller, so that the flow control module 2 dynamically adjusts the gas supply amount Qcool and the supply time tcool based on the control signal.

[0194] In one embodiment, the method further includes: using an electronic expansion valve to dynamically adjust the gas supply quantity Qcool and the supply time tcool based on a control signal.

[0195] In one embodiment, the method further includes: using an electronic multi-way valve to dynamically adjust the gas supply amount Qcool and the supply time tcool based on the control signal. The electronic multi-way valve includes a plurality of input ports and an output port, wherein the plurality of input ports are connected to the vehicle air conditioning system 3, and the output port is connected to the vehicle refrigerator 4.

[0196] In one embodiment, the method further comprises:

[0197]

[0198] e(t) = |Tpred-Tbox|;

[0199] Among them, e(t) is the difference between the real-time temperature Tbox in the car refrigerator and the predicted result Tpred, Kp is the gain coefficient of e(t), Ki is the integral gain coefficient of e(t), Kd is the differential gain coefficient of e(t), and u(t) is the generated control signal.

[0200] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by adding the necessary general hardware platform with the help of software, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0201] An embodiment of the present invention further provides a control system for a vehicle refrigerator, comprising:

[0202] A control device as any of the above;

[0203] The vehicle air conditioning system 3 is used to generate gas, wherein the gas includes cold gas or hot gas;

[0204] A flow control module 2, used for adjusting the flow of gas based on the control device;

[0205] The vehicle refrigerator 4 is used to change the temperature inside the vehicle refrigerator 4 based on the conditioned gas.

[0206] In one implementation, the gas generated by the vehicle air-conditioning system 3 may be the gas directly generated by the vehicle air-conditioning system 3 , or may be the gas refluxed from the vehicle air-conditioning system 3 .

[0207] An embodiment of the present invention further provides a car, comprising any control system as described above.

[0208] An embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0209] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0210] An embodiment of the present invention further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0211] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0212] An embodiment of the present invention further provides a computer program product, including computer instructions, which, when executed by a processor, implement the steps of the method described in each embodiment of the present invention.

[0213] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0214] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0215] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A control device for a vehicle refrigerator, characterized in that: include: A machine learning model module, used to obtain historical data within a historical time period and real-time data at a first moment, wherein the historical data and the real-time data both include: an external environment temperature Text of the vehicle refrigerator, a real-time temperature Tbox inside the vehicle refrigerator, a set target temperature Tset inside the vehicle refrigerator, a vehicle speed Vveh, and an ideal storage temperature Titem of items in the vehicle refrigerator; and, for training the machine learning model of the machine learning model module based on the historical data so that the machine learning model converges; and, used to predict the temperature change in the vehicle refrigerator after the first moment based on the real-time data to obtain a prediction result; and Used to calculate the gas supply quantity Qcool and supply time tcool that need to be adjusted by the flow control module based on the prediction result, so as to control the flow of gas provided by the vehicle air conditioning system to the vehicle refrigerator; a regulating module, configured to enable the flow control module to control the temperature in the vehicle refrigerator to reach the set target temperature Tset in the vehicle refrigerator at a second moment based on the gas supply quantity Qcool and the supply time tcool that need to be adjusted; Wherein, the temperature prediction algorithm for calculating the prediction result is: Tpred=f(Text, Vveh, Tbox, Titem), wherein Tpred is the prediction result, and f is the prediction function of the machine learning model; The algorithm for calculating the gas supply amount Qcool is: Qcool=α*|Tpred-Tbox|, where α is an adjustment factor for optimizing the gas flow rate according to the real-time data and historical data of the vehicle; The algorithm for calculating the supply time tcool is: tcool=β*Qcool, where β is a time adjustment factor used to determine the time required to reach the set target temperature Tset in the vehicle refrigerator; The control device is connected to the vehicle air-conditioning system and the vehicle refrigerator, and the flow control module is connected to the control device, the vehicle air-conditioning system and the vehicle refrigerator.

2. The control device according to claim 1, characterized in that: The adjustment module is also used for: Obtaining the real-time temperature Tbox in the vehicle refrigerator at the third moment, and dynamically adjusting the gas supply amount Qcool and the supply time tcool based on the real-time temperature Tbox in the vehicle refrigerator at the third moment, so that the flow control module controls the temperature in the vehicle refrigerator to reach the set target temperature Tset in the vehicle refrigerator at the second moment; The third moment is any moment between the first moment and the second moment.

3. The control device according to claim 1, characterized in that: The temperature prediction algorithm also includes: Tpred=f(Text,Vveh,Tbox,Titem) =w1*Text+w2*Vveh+w3*Tbox+w4*Titem+b; Among them, w1, w2, w3, and w4 are weight parameters, indicating the influence of each input variable on the prediction result; b is a bias term used to adjust the benchmark of the prediction result.

4. The control device according to claim 1, characterized in that: Also includes: A closed-loop control module, used to calculate the difference between the adjusted real-time temperature Tbox in the vehicle refrigerator and the set target temperature Tset in the vehicle refrigerator; as well as, A control signal is generated based on the difference and a PID controller, so that the flow control module dynamically adjusts the gas supply amount Qcool and the supply time tcool based on the control signal.

5. The control device according to claim 4, characterized in that: The flow control module comprises: The electronic expansion valve is used to dynamically adjust the gas supply quantity Qcool and the supply time tcool based on the control signal.

6. The control device according to claim 4, characterized in that: The flow control module comprises: An electronic multi-way valve comprises a plurality of input ports and output ports, wherein the plurality of input ports are connected to the vehicle air conditioning system, and the output port is connected to the vehicle refrigerator, so as to control the opening or closing of the plurality of input ports based on the control signal.

7. The control device according to claim 4, characterized in that: The control signal generation algorithm includes: e(t) = |Tpred-Tbox|; Among them, e(t) is the difference between the real-time temperature Tbox in the car refrigerator and the predicted result Tpred, Kp is the gain coefficient of e(t), Ki is the integral gain coefficient of e(t), Kd is the differential gain coefficient of e(t), and u(t) is the generated control signal.

8. A method for controlling a vehicle refrigerator, characterized in that: The method is applied to a control device according to any one of claims 1 to 7, comprising: Acquire historical data within a historical time period to train the machine learning model of the machine learning model module based on the historical data so as to converge the machine learning model; Acquire real-time data at a first moment, so as to predict the temperature change in the vehicle refrigerator after the first moment based on the real-time data, so as to obtain a prediction result; wherein the historical data and the real-time data both include: the external environment temperature Text of the vehicle refrigerator, the real-time temperature Tbox in the vehicle refrigerator, the set target temperature Tset in the vehicle refrigerator, the vehicle driving speed Vveh, and the ideal storage temperature Titem of the items in the vehicle refrigerator; Based on the prediction result, the gas supply quantity Qcool and the supply time tcool that the flow control module needs to adjust are calculated to control the flow of gas provided by the vehicle air conditioning system to the vehicle refrigerator; Based on the gas supply quantity Qcool and the supply time tcool that need to be adjusted, the flow control module controls the temperature in the vehicle refrigerator to reach the set target temperature Tset in the vehicle refrigerator at the second moment; Wherein, the temperature prediction algorithm for calculating the prediction result is: Tpred=f(Text, Vveh, Tbox, Titem), wherein Tpred is the prediction result, and f is the prediction function of the machine learning model; The algorithm for calculating the gas supply amount Qcool is: Qcool=α*|Tpred-Tbox|, where α is an adjustment factor for optimizing the gas flow rate according to the real-time data and historical data of the vehicle; The algorithm for calculating the supply time tcool is: tcool=β*Qcool, where β is a time adjustment factor used to determine the time required to reach the set target temperature Tset in the vehicle refrigerator; The control device is connected to the vehicle air-conditioning system and the vehicle refrigerator, and the flow control module is connected to the control device, the vehicle air-conditioning system and the vehicle refrigerator.

9. A control system for a vehicle refrigerator, characterized in that: include: A control device as claimed in any one of claims 1 to 7; A vehicle air conditioning system for generating gas, wherein the gas includes cold gas or hot gas; a flow control module, used for adjusting the flow of the gas based on the control device; A vehicle refrigerator is used to change the temperature inside the vehicle refrigerator based on the conditioned gas.

10. An automobile, characterized in that: Comprising a control system as claimed in claim 9.

Images