Energy-saving temperature control method for vehicle-mounted air conditioner of dormitory van

Abstract

According to the energy-saving temperature control method for the vehicle-mounted air conditioner of the dormitory van, by introducing personnel existence state perception and judgment, a first energy-saving temperature control strategy can be automatically started when it is confirmed that a compartment is in an unmanned state, invalid energy consumption of the air conditioner in the unmanned period is avoided, remarkable energy conservation and consumption reduction are achieved, and the energy-saving temperature control method for the vehicle-mounted air conditioner of the dormitory van is achieved. And the cruising ability of the new energy boarding car is particularly improved. And when it is judged that the passenger is in the manned state, the second comfortable temperature control strategy is intelligently started, it is ensured that an accurate and comfortable temperature environment is provided for the passenger, and therefore the energy-saving effect is maximized on the premise that comfort is guaranteed. According to the whole method, the operation mode of the vehicle-mounted air conditioner is managed, the optimal balance of energy conservation and comfort is achieved, and then the technical problem that in the prior art, only the single physical parameter of the temperature is concerned, and the existing state of people in the vehicle cannot be sensed is solved.

Description

Technical Field

[0001] This invention relates to the field of vehicle air conditioning temperature control technology, specifically to an energy-saving temperature control method for a campervan's air conditioning system. Background Technology

[0002] Currently, most campervan air conditioning systems directly borrow the control logic from traditional automotive or household air conditioners. Their core principle is to use a user-set target temperature, feedback from an in-cabin temperature sensor, and simple on / off control or proportional-integral-derivative (PID) control algorithms to start / stop the compressor or adjust the fan speed, thereby maintaining the cabin temperature near the set value.

[0003] However, current technologies only focus on the single physical parameter of temperature and cannot detect the presence of people inside the vehicle. In real-world campervan use cases, occupants often temporarily leave the vehicle for outdoor activities, leaving the interior unoccupied. Existing control systems cannot detect this, and the air conditioning system continues to run to maintain the preset temperature, resulting in significant energy or fuel consumption in ineffective temperature control. For parking air conditioners that rely on limited onboard battery power, this severely shortens outdoor driving time and may damage battery life; for driving air conditioners, it increases unnecessary engine load and fuel consumption. Summary of the Invention

[0004] The purpose of this invention is to provide an energy-saving temperature control method for the air conditioning of a campervan, so as to solve the technical problem that the existing technology only focuses on the single physical parameter of temperature and cannot sense the presence of people inside the vehicle.

[0005] The technical solution of this invention is implemented as follows:

[0006] A method for energy-saving temperature control of a campervan's onboard air conditioning system includes:

[0007] Step S1: Monitor the presence status of people in the camper car in real time. The presence status of people includes at least an unoccupied state and an occupied state.

[0008] Step S2: Select and switch the corresponding air conditioning temperature control mode according to the status of the personnel;

[0009] Step S3: Using a preset personnel density model, when the area is unoccupied, activate the first energy-saving temperature control strategy; when the area is occupied, activate the second comfort temperature control strategy.

[0010] Step S4: Based on the enabled air conditioning temperature control mode, generate air conditioning control commands to control the operation of the vehicle air conditioning.

[0011] A further technical solution is that step S1 specifically includes:

[0012] Step S11: Acquire raw sensor signals reflecting the presence or activity of living organisms in real time by using at least one non-visual sensor installed in the camper car compartment;

[0013] Step S12: Filter and extract features from the original sensing signal to obtain at least one feature related to the presence of people;

[0014] Step S13: Compare and analyze the at least one feature information with a preset judgment threshold or feature model to determine whether the carriage is currently occupied or unoccupied.

[0015] Step S14: Convert the determination result of whether the person is in the state or the person is not in the state into a status signal that can be recognized by the air conditioning control system and output it.

[0016] A further technical solution is that step S13 specifically includes:

[0017] Step S131: Receive at least one of the aforementioned feature information; the feature information includes the intensity of vital micro-motion characteristics, heat source distribution characteristics, or effective motion trajectory;

[0018] Step S132: According to the system configuration, call the preset judgment threshold or pre-trained feature model corresponding to the feature information;

[0019] Step S133: Compare the feature information with the preset judgment threshold, or input it into the pre-trained feature model for calculation to obtain one or more state judgment values;

[0020] Step S134: Based on the one or more state determination values, determine whether the carriage is currently occupied or unoccupied according to the preset comprehensive decision logic.

[0021] Step S135: Output the determination result of whether the state is occupied or unoccupied to the decision module.

[0022] A further technical solution is that step S2 includes:

[0023] Step S21: Obtain the status determination signal output by the monitoring module for the presence of personnel, indicating whether the carriage is currently occupied or unoccupied.

[0024] Step S22: Based on the received state determination signal, map and select the corresponding target temperature control strategy from the pre-stored temperature control strategy library;

[0025] Step S23: Based on the selected target temperature control strategy, determine and generate a set of corresponding air conditioning mode control parameters; wherein, the control parameters include at least one or more of the following: target temperature setpoint, temperature control dead zone range, and fan operation mode;

[0026] Step S24: Based on the air conditioning mode control parameters, form a set of control instructions that can be parsed and executed by the vehicle air conditioning actuator, and output the set of control instructions to the air conditioning control mode.

[0027] A further technical solution is that step S22 specifically includes:

[0028] Step S221: Receive the state determination signal and parse the state determination signal to represent the carriage state as either the occupied state or the unoccupied state.

[0029] Step S222: Input the parsed state determination signal into a preset strategy mapping logic unit; the strategy mapping logic unit stores the mapping relationship between the first comfort strategy and the second energy-saving strategy, and outputs a candidate temperature control strategy identifier based on the mapping relationship;

[0030] Step S223: Based on the candidate temperature control strategy identifier, search in the pre-stored temperature control strategy library and call the set of strategy parameters bound to the candidate temperature control strategy identifier.

[0031] A further technical solution is that step S223 specifically includes:

[0032] Step S2231: Extract the candidate temperature control strategy identifier as a search key;

[0033] Step S2232: Based on the search key, generate a search request pointing to the pre-stored temperature control strategy library;

[0034] Step S2233: Send the retrieval request to the temperature control strategy library, and traverse and match the storage entries that are consistent with or associated with the retrieval key in the strategy library;

[0035] Step S2234: When a match is successful, read and return the set of strategy parameters corresponding to the stored entry from the temperature control strategy library;

[0036] Step S2235: Logically bind the returned set of strategy parameters with the candidate temperature control strategy identifier.

[0037] A further technical solution is that step S3 specifically includes:

[0038] Step S31: Determine the calculation path of the preset personnel density model that needs to be called;

[0039] Step S32: Along the determined calculation path, call and execute the preset personnel density model, take the personnel presence status signal as input, calculate and output the density value of the personnel density level in the carriage;

[0040] Step S33: Compare the density value with a preset strategy matching threshold: if the density value falls within the first threshold range for unoccupied areas, it is matched to the first energy-saving temperature control strategy; if the density value falls within the second threshold range for occupied areas, it is matched to the second comfort temperature control strategy.

[0041] Step S34: Generate a set of control parameters for direct parsing based on the matched first energy-saving temperature control strategy or second comfort temperature control strategy.

[0042] A further technical solution is that step S31 specifically includes:

[0043] Step S311: Analyze the personnel presence status signal to distinguish the specific status category it represents. The status category includes the unmanned state and the manned state.

[0044] Step S312: Based on the state category, map it to the corresponding calculation logic branch in the preset personnel density model, wherein the unmanned state is mapped to the first calculation logic branch, and the manned state is mapped to the second calculation logic branch.

[0045] A further technical solution is that step S32 specifically includes:

[0046] Step S321: Format the personnel presence status signal into the input data format required by the preset personnel density model;

[0047] Step S322: Load the model calculation logic corresponding to the calculation path and pass in the input data;

[0048] Step S323: Run the model calculation logic to perform calculations on the input data and generate intermediate results representing the population density level;

[0049] Step S324: Process the intermediate results and output the density value.

[0050] A further technical solution is that the preset personnel density model includes a Gaussian process regression model.

[0051] The beneficial effects of this invention are as follows:

[0052] By introducing the perception and judgment of the presence of occupants, the system automatically activates the first energy-saving temperature control strategy when the vehicle is confirmed to be unoccupied, avoiding ineffective energy consumption of the air conditioner during unoccupied periods and achieving significant energy savings, especially beneficial for improving the range of new energy camping vehicles. When the system determines that someone is present, it intelligently activates the second comfort temperature control strategy to ensure a precise and comfortable temperature environment for passengers, thereby maximizing energy savings while ensuring comfort. The entire method manages the operating mode of the vehicle's air conditioning, achieving an optimal balance between energy saving and comfort, and thus solving the technical problem of existing technologies that only focus on the single physical parameter of temperature and cannot perceive the presence of occupants inside the vehicle. Attached Figure Description

[0053] Figure 1 This is a flowchart illustrating the steps of an energy-saving temperature control method for an onboard air conditioner in a campervan according to the present invention.

[0054] Figure 2 This is a flowchart of step S1 of a campervan air conditioning energy-saving temperature control method according to the present invention;

[0055] Figure 3 This is a flowchart of step S2 of the energy-saving temperature control method for an onboard air conditioner in a campervan according to the present invention;

[0056] Figure 4 This is a flowchart of step S3 of a campervan air conditioning energy-saving temperature control method according to the present invention. Detailed Implementation

[0057] To better understand the technical content of this invention, specific embodiments are provided below, and the invention will be further described in conjunction with the accompanying drawings.

[0058] See Figures 1 to 2 This invention provides an energy-saving temperature control method for onboard air conditioning in a campervan, comprising:

[0059] Step S1: Monitor the presence status of people in the camper car in real time. The presence status includes at least an unoccupied state and an occupied state.

[0060] Step S2: Select and switch the corresponding air conditioning temperature control mode according to the presence of people;

[0061] Step S3: Using a preset personnel density model, activate the first energy-saving temperature control strategy when the area is unoccupied; activate the second comfort temperature control strategy when the area is occupied.

[0062] Step S4: Based on the enabled air conditioning temperature control mode, generate air conditioning control commands to control the operation of the vehicle air conditioning.

[0063] Specifically, the vehicle's onboard sensor system uses real-time sensing and analysis of the environmental information inside the passenger compartment to accurately determine whether the compartment is unoccupied or occupied. This determination is then used as a decision input to trigger a switch in the control logic: when the compartment is "unoccupied," an air conditioning temperature control mode designed to reduce standby power consumption is selected and switched; when the compartment is "occupied," another air conditioning temperature control mode is switched to prioritize passenger comfort.

[0064] By invoking a preset personnel density model, this model can analyze and calculate sensor data (such as heat source distribution and micro-motion signals), thereby providing data support for determining the presence of people or classifying the status. Based on this, when the absence of people is confirmed, a first energy-saving temperature control strategy is activated. This strategy may include relaxing the allowable temperature fluctuation range, shutting off ventilation in some areas, or putting the air conditioner into a low-power intermittent operation mode. When the presence of people is confirmed, a second comfort temperature control strategy is activated. This strategy typically strictly controls the interior temperature within a preset comfort range and maintains necessary air circulation. Finally, the specific temperature control strategies activated above are converted into control commands (such as set temperature, fan speed, and operating mode signals) that can be directly recognized and executed by the vehicle's air conditioning actuators (such as compressors, fans, and dampers). This completes the control of the vehicle's air conditioning operation, ultimately ensuring comfort during periods of presence while minimizing energy waste during periods of absence.

[0065] This invention introduces the perception and judgment of the presence of occupants. When the vehicle is confirmed to be unoccupied, it automatically activates a first energy-saving temperature control strategy to avoid ineffective energy consumption by the air conditioner during unoccupied periods, achieving significant energy savings, especially beneficial for improving the range of new energy camping vehicles. When the presence of occupants is detected, a second comfort temperature control strategy is intelligently activated to ensure a precise and comfortable temperature environment for passengers, thereby maximizing energy savings while ensuring comfort. The entire method manages the operating mode of the vehicle's air conditioning, achieving an optimal balance between energy saving and comfort, thus solving the technical problem of existing technologies that only focus on the single physical parameter of temperature and cannot perceive the presence of occupants inside the vehicle.

[0066] In this embodiment of the invention, step S1 specifically includes:

[0067] Step S11: Acquire raw sensor signals reflecting the presence or activity of living organisms in real time by using at least one non-visual sensor installed in the camper car compartment;

[0068] Step S12: Filter and extract features from the original sensing signal to obtain at least one feature related to the presence of people;

[0069] Step S13: Compare and analyze at least one feature with a preset judgment threshold or feature model to determine whether the carriage is currently occupied or unoccupied.

[0070] Step S14: Convert the determination result of whether the person is in the state or the person is not in the state into a status signal that can be recognized by the air conditioning control system and output it.

[0071] Specifically, continuous scanning is performed by at least one non-visual sensor (such as millimeter-wave radar or an infrared sensor array) deployed within the vehicle compartment to capture raw physical signals containing micro-movements, thermal radiation, or motion trajectories of living beings, achieving non-destructive acquisition of environmental information. Next, the raw signals are preprocessed (e.g., noise reduction filtering) to extract feature information indicating the presence or absence of personnel from the time, frequency, or spatial domains (e.g., signal energy in a specific frequency band, heat source contours, or motion continuity indicators). Subsequently, the extracted feature information is compared with pre-set judgment thresholds through experimentation or learning, or input into a pre-trained binary classification feature model (such as a support vector machine or lightweight neural network) for inference analysis, thereby outputting a deterministic logical judgment regarding whether the vehicle compartment is currently occupied or unoccupied. Finally, this logical judgment is converted into a status signal conforming to the internal communication protocol of the vehicle's air conditioning control system (such as CAN bus messages or specific level signals) and output.

[0072] Furthermore, step S13 specifically includes:

[0073] Step S131: Receive at least one feature information; the feature information includes the intensity of micro-movement characteristics, heat source distribution characteristics, or effective motion trajectory;

[0074] Step S132: According to the system configuration, call the preset judgment threshold or pre-trained feature model corresponding to the feature information;

[0075] Step S133: Compare the feature information with the preset judgment threshold, or input it into the pre-trained feature model for calculation to obtain one or more state judgment values;

[0076] Step S134: Based on one or more state determination values, determine whether the carriage is currently occupied or unoccupied according to the preset comprehensive decision logic.

[0077] Step S135: Output the determination result of whether the state is occupied or unoccupied to the decision module.

[0078] Specifically, the system receives at least one digital feature, which is a key indicator reflecting the possibility of a person's presence, such as the intensity of vital signs, the spatial distribution characteristics of heat sources, or an effective movement trajectory after trajectory analysis. Based on the current configuration, it invokes the corresponding decision rule base: if a threshold judgment rule is used, it invokes a preset judgment threshold bound to the specific feature; if a model judgment rule is used, it invokes a feature model (such as a classifier) ​​that has been trained offline. Depending on the type of rule invoked, the received feature information is numerically compared with the preset threshold, or it is fed as an input vector into the feature model for forward inference calculation, thereby generating one or more intermediate state judgment values ​​(e.g., independent judgment results from each sensor or classification probabilities output by the model). These one or more judgment values ​​are input into a preset comprehensive decision logic (e.g., an "AND" logic requires all independent evidence to support the same conclusion, an "OR" logic allows a single strong piece of evidence to make a judgment, or a weighted voting mechanism). Through the integrated analysis of this logic, a binary conclusion is finally made: the carriage is currently occupied or unoccupied. This final state judgment result, after comprehensive evaluation, is output to the air conditioning temperature control mode decision module.

[0079] In this embodiment of the invention, step S2 includes:

[0080] Step S21: Obtain the status determination signal output by the personnel presence monitoring module, indicating whether the carriage is currently occupied or unoccupied.

[0081] Step S22: Based on the received status determination signal, map and select the corresponding target temperature control strategy from the pre-stored temperature control strategy library;

[0082] Step S23: Based on the selected target temperature control strategy, determine and generate a set of corresponding air conditioning mode control parameters; wherein, the control parameters include at least one or more of the following: target temperature setpoint, temperature control dead zone range, and fan operation mode;

[0083] Step S24: Based on the air conditioning mode control parameters, form a set of control instructions that can be parsed and executed by the vehicle air conditioning actuator, and output the set of control instructions to the air conditioning control mode.

[0084] Specifically, the system receives logical signals indicating the real-time status of the vehicle interior (clearly indicating whether the vehicle is occupied or unoccupied) from front-end personnel presence monitoring. Using these presence signals as an index, it queries a pre-stored temperature control strategy library. This library pre-establishes a mapping between status and strategy (e.g., unoccupied status maps to "energy-saving strategy A," and occupied status maps to "comfort strategy B"), and retrieves and invokes the corresponding target temperature control strategy accordingly. Based on the rules defined by the selected target temperature control strategy, a set of air conditioning mode control parameters is determined and generated. These parameters include at least the core target temperature setpoint, the allowable temperature fluctuation dead zone, and the specific operating mode of the fan (e.g., off, low speed, or automatic). The control parameter set is encapsulated into a set of control instructions (e.g., specific CAN bus messages or PWM signals) that can be directly parsed and executed by the vehicle's air conditioning controller or actuator, according to the communication protocol and data format required by the controller. This instruction is then output to the air conditioning control module, driving the air conditioning system to operate in a mode matching the personnel's status, achieving a dynamic balance between energy saving and comfort.

[0085] Furthermore, step S22 specifically includes:

[0086] Step S221: Receive the status determination signal and parse the status determination signal to represent the carriage status as occupied or unoccupied.

[0087] Step S222: Input the parsed state determination signal into the preset strategy mapping logic unit; the strategy mapping logic unit stores the mapping relationship between the first comfort strategy and the second energy-saving strategy, and outputs the candidate temperature control strategy identifier based on the mapping relationship;

[0088] Step S223: Based on the candidate temperature control strategy identifier, search in the pre-stored temperature control strategy library and call the set of strategy parameters bound to the candidate temperature control strategy identifier.

[0089] Specifically, upon receiving a status determination signal, the system analyzes its specific encoding or logic level to explicitly convert it into a status identifier indicating whether the carriage is currently occupied or unoccupied. This status identifier is then submitted as input to a pre-defined strategy mapping logic unit (which can be represented as a lookup table or a conditional judgment function). This unit internally establishes mapping rules between status and strategy types (i.e., occupied status maps to a first comfort strategy, and unoccupied status maps to a second energy-saving strategy), and outputs a candidate temperature control strategy identifier corresponding to the target strategy type based on these rules. Using this candidate temperature control strategy identifier as a search key, a query is performed in a pre-stored temperature control strategy library (whose storage structure can be a database or a configuration file). When a unique entry is successfully matched, the system calls and reads the complete set of strategy parameters permanently bound to that identifier, containing all specific control instructions and setpoints. This completes the conversion from logical status to executable strategy parameters, providing a direct parameter basis for generating the final control instructions.

[0090] Furthermore, step S223 specifically includes:

[0091] Step S2231: Extract the candidate temperature control strategy identifier as the search key;

[0092] Step S2232: Based on the search key, generate a search request pointing to the pre-stored temperature control strategy library;

[0093] Step S2233: Send the retrieval request to the temperature control strategy library, and traverse and match the storage entries that are consistent with or related to the retrieval key in the strategy library.

[0094] Step S2234: When a match is successful, read and return the set of strategy parameters corresponding to the stored entry from the temperature control strategy library;

[0095] Step S2235: Logically bind the returned set of strategy parameters with the candidate temperature control strategy identifier.

[0096] Specifically, the candidate temperature control strategy identifier is extracted as the unique key index. Based on this search key, a data retrieval request is constructed according to a predefined communication or query protocol. This request explicitly points to the pre-stored temperature control strategy library. The generated search request is sent to the strategy library. Upon receiving the request, the strategy library traverses and compares its internal data structures (such as hash tables or database tables) to find specific stored entries whose primary key or related fields are completely consistent with the search key or have a strong correlation mapping relationship. When a corresponding entry is successfully matched, its strategy parameter set (including all specific set values ​​such as temperature and wind speed) is read from the stored entry, and this set of data is returned as the query result. The returned set of strategy parameters is logically associated and bound with the candidate temperature control strategy identifier that initially initiated the search, forming a complete strategy object with a strategy type identifier and all execution parameters.

[0097] In this embodiment of the invention, step S3 specifically includes:

[0098] Step S31: Determine the calculation path of the preset personnel density model that needs to be called;

[0099] Step S32: Along the determined calculation path, call and execute the preset personnel density model, take the personnel presence status signal as input, calculate and output the density value of the personnel density level in the carriage;

[0100] Step S33: Compare the density value with the preset strategy matching threshold: If the density value falls within the first threshold range for unoccupied areas, it is matched to the first energy-saving temperature control strategy; if the density value falls within the second threshold range for occupied areas, it is matched to the second comfort temperature control strategy.

[0101] Step S34: Generate a set of control parameters for direct parsing based on the matched first energy-saving temperature control strategy or second comfort temperature control strategy.

[0102] Specifically, based on the preliminary state category of the received personnel presence status signal (e.g., "occupied" or "unoccupied"), the corresponding calculation logic branch or path is determined within the preset personnel density model to prepare for the model's operation. Following this determined path, the personnel density model is invoked and executed, using the status signal (which may include raw sensor features or intermediate judgment results) as input data. The personnel density model processes and calculates through its internal algorithm, ultimately outputting a continuous or discrete density value representing the real-time personnel density level inside the carriage. The core strategy matching logic is then executed: the calculated density value is compared and judged against a pre-set strategy matching threshold stored in the system. If the density value falls within the first threshold range representing "unoccupied" (e.g., equal to zero or below a certain extremely low threshold), it is matched to the first energy-saving temperature control strategy; if the density value falls within the second threshold range representing "occupied" (e.g., greater than zero or above a certain activation threshold), it is matched to the second comfort temperature control strategy. Based on the matching results above, a set of control parameters (e.g., including specific set temperature, compressor working mode, fan speed level, etc.) that are completely corresponding to the selected strategy (first energy saving or second comfort temperature control strategy) and can be directly parsed and executed by the air conditioner's underlying controller are obtained from the strategy library or generated in real time, thereby completing the intelligent decision-making and conversion process from status signals to final execution instructions.

[0103] Furthermore, step S31 specifically includes:

[0104] Step S311: Analyze the presence status signal of personnel to distinguish the specific status category it represents. The status category includes unmanned state and manned state.

[0105] Step S312: Based on the state category, map to the corresponding calculation logic branch in the preset personnel density model, where the unmanned state is mapped to the first calculation logic branch, and the manned state is mapped to the second calculation logic branch.

[0106] Specifically, the input personnel presence status signal is analyzed to identify and decode the specific logical information carried by the signal, thereby clearly distinguishing two discrete state categories: unoccupied or occupied. This analysis is a prerequisite for subsequent targeted calculations. Based on the identified state category, a predefined mapping operation is performed: the state category is used as an input condition and matched with different computational logic branches within the pre-defined personnel density model.

[0107] Specifically, the mapping rule stipulates that when the state category is "unoccupied," it is mapped to the first computational logic branch of the model; when the state category is "occupied," it is mapped to the second computational logic branch of the model. Thus, based on different macroscopic human presence scenarios, the algorithm execution path is specified for density calculation, ensuring the efficiency and relevance of the model's computation.

[0108] Furthermore, step S32 specifically includes:

[0109] Step S321: Format the personnel presence status signal into the input data format required by the preset personnel density model;

[0110] Step S322: Load the model calculation logic corresponding to the calculation path and pass in the input data;

[0111] Step S323: Run the model calculation logic to perform calculations on the input data and generate intermediate results representing the population density level;

[0112] Step S324: Process the intermediate results and output the density values.

[0113] Specifically, the original communication protocol format or logical state is converted and reconstructed into a specific structured input data format required by the internal algorithm of the preset personnel density model. Based on the determined specific computation path (e.g., the first or second computation logic branch), the instantiated model computation logic (i.e., a specific algorithm program or function) corresponding to that path is loaded from memory, and the formatted input data is passed as parameters to this logic unit. The loaded model computation logic is run, which processes and calculates the input data based on its built-in algorithm, generating intermediate results (e.g., probability values, scores, or unstandardized estimates) of the personnel density level inside the carriage. Necessary post-processing is performed on these intermediate results, such as through standardization, range mapping, or threshold judgment, to convert them into density values ​​with clear physical or logical meaning (e.g., a continuous scale between 0 and 1, or discrete levels such as "low," "medium," and "high"), and the final density value is output for use in subsequent strategy matching steps, thus completing the entire computation chain from the original state signal to the quantified density index.

[0114] Furthermore, the preset personnel density model includes a Gaussian process regression model.

[0115] Specifically, by modeling the mapping relationship between sensor features and personnel density as a set of stochastic processes, accurate inference and uncertainty quantification of personnel density levels inside the carriage can be achieved.

[0116] Specifically, during the offline training phase, the Gaussian process regression model learns using multiple sets of historically collected sample data. Each set of sample data includes input feature vectors extracted from non-visual sensors (such as millimeter-wave radar and infrared sensors) (e.g., the spectral energy of vital micro-motion signals, the distribution area of ​​heat sources, and the length of movement trajectories) and corresponding real human density values ​​obtained through calibration (e.g., 0, 1, 2, ... people). By learning from these samples, the Gaussian process regression model determines a Gaussian process prior distribution defined by the mean function and the covariance function (or kernel function). This distribution describes the possible density output values ​​and their confidence levels under any input features.

[0117] During the real-time inference phase, the Gaussian process regression model uses historical data or real-time sensor feature information (such as the micro-motion energy spectrum of millimeter-wave radar or the thermal radiation intensity of a specific area in infrared thermal imaging) as training input to learn the nonlinear mapping relationship from multi-dimensional features to continuous personnel density values ​​and construct a probabilistic generation model. During the execution phase, the Gaussian process regression model takes the sensor feature vector at the current moment as input and, through Gaussian process regression calculation, outputs not only a continuous predicted mean of personnel density but, more importantly, also the confidence interval or variance of this predicted value. When the density prediction mean output by the Gaussian process regression model increases, and its lower confidence interval exceeds the preset threshold for occupant activation, it indicates a high degree of confidence in the increase in personnel density within the carriage, thus determining that the cooling effect needs to be enhanced. If the current temperature has not reached the lower limit of the comfort temperature control strategy, the required cooling capacity adjustment will be further calculated: on the one hand, based on the increment of the predicted mean density, and on the other hand, considering the size of the confidence interval. If the confidence interval is narrow (strong prediction certainty), the compressor power will be increased proportionally or the target temperature will be lowered based on the increment of the mean. If the confidence interval is wide (large prediction uncertainty), a more conservative adjustment strategy will be adopted, such as only slightly increasing the cooling capacity output and continuously monitoring subsequent data changes to avoid energy waste or decreased comfort due to misjudgment. In this way, the Gaussian process regression model achieves the estimation of personnel density and provides a robust and adaptive decision-making basis for whether to increase the cooling capacity.

[0118] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for energy-saving temperature control of an onboard air conditioner in a campervan, characterized in that, include: Step S1: Monitor the presence status of personnel in the camper car in real time. The presence status of personnel includes at least an unoccupied state and an occupied state. Step S2: Select and switch the corresponding air conditioning temperature control mode according to the status of the personnel; Step S3: Using a preset personnel density model, when the area is unoccupied, activate the first energy-saving temperature control strategy; when the area is occupied, activate the second comfort temperature control strategy. Step S4: Based on the enabled air conditioning temperature control mode, generate air conditioning control commands to control the operation of the vehicle air conditioning.

2. The energy-saving temperature control method for a campervan's onboard air conditioning according to claim 1, characterized in that, Step S1 specifically includes: Step S11: Acquire raw sensor signals reflecting the presence or activity of living organisms in real time by using at least one non-visual sensor installed in the camper car compartment; Step S12: Filter and extract features from the original sensing signal to obtain at least one feature related to the presence of people; Step S13: Compare and analyze the at least one feature information with a preset judgment threshold or feature model to determine whether the carriage is currently occupied or unoccupied. Step S14: Convert the determination result of whether the person is in the state or the person is not in the state into a status signal that can be recognized by the air conditioning control system and output it.

3. The energy-saving temperature control method for a campervan's onboard air conditioning according to claim 2, characterized in that, Step S13 specifically includes: Step S131: Receive at least one of the aforementioned feature information; the feature information includes the intensity of vital micro-motion characteristics, heat source distribution characteristics, or effective motion trajectory; Step S132: According to the system configuration, call the preset judgment threshold or pre-trained feature model corresponding to the feature information; Step S133: Compare the feature information with the preset judgment threshold, or input it into the pre-trained feature model for calculation to obtain one or more state judgment values; Step S134: Based on the one or more state determination values, determine whether the carriage is currently occupied or unoccupied according to the preset comprehensive decision logic. Step S135: Output the determination result of whether the state is occupied or unoccupied to the decision module.

4. The energy-saving temperature control method for onboard air conditioning in a camping vehicle according to claim 1, characterized in that, Step S2 includes: Step S21: Obtain the status determination signal output by the monitoring module for the presence of personnel, indicating whether the carriage is currently occupied or unoccupied. Step S22: Based on the received state determination signal, map and select the corresponding target temperature control strategy from the pre-stored temperature control strategy library; Step S23: Based on the selected target temperature control strategy, determine and generate a set of corresponding air conditioning mode control parameters; wherein, the control parameters include at least one or more of the following: target temperature setpoint, temperature control dead zone range, and fan operation mode; Step S24: Based on the air conditioning mode control parameters, form a set of control instructions that can be parsed and executed by the vehicle air conditioning actuator, and output the set of control instructions to the air conditioning control mode.

5. The energy-saving temperature control method for onboard air conditioning in a camping vehicle according to claim 4, characterized in that, Step S22 specifically includes: Step S221: Receive the state determination signal and parse the state determination signal to represent the carriage state as either the occupied state or the unoccupied state. Step S222: Input the parsed state determination signal into a preset strategy mapping logic unit; the strategy mapping logic unit stores the mapping relationship between the first comfort strategy and the second energy-saving strategy, and outputs a candidate temperature control strategy identifier based on the mapping relationship; Step S223: Based on the candidate temperature control strategy identifier, search in the pre-stored temperature control strategy library and call the set of strategy parameters bound to the candidate temperature control strategy identifier.

6. The energy-saving temperature control method for a campervan's onboard air conditioning according to claim 5, characterized in that, Step S223 specifically includes: Step S2231: Extract the candidate temperature control strategy identifier as a search key; Step S2232: Based on the search key, generate a search request pointing to the pre-stored temperature control strategy library; Step S2233: Send the retrieval request to the temperature control strategy library, and traverse and match the storage entries that are consistent with or associated with the retrieval key in the strategy library; Step S2234: When a match is successful, read and return the set of strategy parameters corresponding to the stored entry from the temperature control strategy library; Step S2235: Logically bind the returned set of strategy parameters with the candidate temperature control strategy identifier.

7. The energy-saving temperature control method for a campervan's onboard air conditioning according to claim 1, characterized in that, Step S3 specifically includes: Step S31: Determine the calculation path of the preset personnel density model that needs to be called; Step S32: Along the determined calculation path, call and execute the preset personnel density model, take the personnel presence status signal as input, calculate and output the density value of the personnel density level in the carriage; Step S33: Compare the density value with a preset strategy matching threshold. If the density value falls within the first threshold range for unoccupied areas, it is matched to the first energy-saving temperature control strategy; if the density value falls within the second threshold range for occupied areas, it is matched to the second comfort temperature control strategy. Step S34: Generate a set of control parameters for direct parsing based on the matched first energy-saving temperature control strategy or second comfort temperature control strategy.

8. The energy-saving temperature control method for a campervan's onboard air conditioning according to claim 7, characterized in that, Step S31 specifically includes: Step S311: Analyze the personnel presence status signal to distinguish the specific status category it represents. The status category includes the unmanned state and the manned state. Step S312: Based on the state category, map it to the corresponding calculation logic branch in the preset personnel density model, wherein the unmanned state is mapped to the first calculation logic branch, and the manned state is mapped to the second calculation logic branch.

9. A method for energy-saving temperature control of a campervan's onboard air conditioning according to claim 7, characterized in that, Step S32 specifically includes: Step S321: Format the personnel presence status signal into the input data format required by the preset personnel density model; Step S322: Load the model calculation logic corresponding to the calculation path and pass in the input data; Step S323: Run the model calculation logic to perform calculations on the input data and generate intermediate results representing the population density level; Step S324: Process the intermediate results and output the density value.

10. A method for energy-saving temperature control of a campervan's onboard air conditioning according to claim 7, characterized in that, The preset personnel density model includes a Gaussian process regression model.

Images