Airplane ground air conditioning energy-saving control method combined with environmental parameter prediction
By combining environmental parameter prediction and aircraft characteristics to optimize air conditioning control, the problems of low control accuracy and high energy consumption of aircraft ground air conditioning systems have been solved, achieving precise comfort control and energy-saving goals, and improving passenger experience and energy efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- JIANGSU SAFE AVIATION TECH CO LTD
- Filing Date
- 2025-09-25
- Publication Date
- 2026-06-26
AI Technical Summary
The existing aircraft ground air conditioning control system has low control precision, poor passenger comfort, high energy consumption, and lacks linkage with flight dynamic information, resulting in delayed or excessive environmental preparation and making it difficult to achieve energy-saving goals.
By acquiring aircraft characteristics and historical environmental parameters, an environmental parameter prediction agent is used to predict future environmental changes. Combined with aircraft landing time error compensation and internal thermal characteristic analysis, air conditioning control parameters are optimized, and a multi-objective optimization algorithm is used to balance comfort and energy consumption.
It achieves forward-looking and precise control of the air conditioning system, overcomes the uncertainty of flight schedules, improves passenger comfort, and significantly reduces energy consumption, thus having significant economic benefits and environmental value.
Smart Images
Figure CN121224993B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of aircraft ground support equipment technology, and more specifically to an energy-saving control method for aircraft ground air conditioning based on environmental parameter prediction. Background Technology
[0002] In aircraft ground handling procedures, the process of passengers being transferred between the terminal and the aircraft via shuttle corridors or shuttle buses is a key part of the ground service experience, and providing passengers with a comfortable temperature environment during this period is crucial. Currently, the air conditioning control systems in shuttle corridors or shuttle buses mostly adopt temperature control strategies based on fixed schedules or roughly set according to local climate, such as uniformly setting a low-temperature cooling mode in summer and a fixed heating mode in winter.
[0003] However, this control method suffers from low control precision and poor passenger comfort. Furthermore, to cope with potential extreme temperatures, it often operates at high power for extended periods, lacking intelligent adjustment mechanisms and resulting in unnecessary energy consumption. In addition, because the current control system is an independent closed loop, it is not linked to flight dynamics information, leading to delayed or excessive environmental preparation. This makes it difficult to achieve energy-saving goals while ensuring passenger comfort, lacking intelligence, foresight, and adaptability. Summary of the Invention
[0004] This invention addresses the technical problems of low control precision, high energy consumption, and lack of intelligent linkage with flight dynamic information in the existing air conditioning control systems for shuttle channels or shuttle buses, which lead to poor passenger comfort and serious energy waste. It provides an energy-saving control method for aircraft ground air conditioning that combines environmental parameter prediction.
[0005] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:
[0006] In a first aspect, the present invention provides an energy-saving control method for aircraft ground air conditioning that incorporates environmental parameter prediction, comprising:
[0007] The aircraft characteristics of the target aircraft are obtained, and when the expected landing time of the target aircraft is less than the landing time threshold, the historical environmental parameter sequence of the ground area within a preset time range is obtained.
[0008] Based on the historical environmental parameter sequence, environmental parameter prediction is performed to obtain a predicted environmental parameter sequence within a preset time range in the future;
[0009] Based on the aircraft characteristics, the aircraft landing time error is classified to obtain the error coefficient, and the expected landing time is compensated to obtain the expected landing time interval. The landing environmental parameter sequence is obtained by indexing within the predicted environmental parameter sequence.
[0010] Based on the aircraft characteristics, the internal thermal characteristic range of the aircraft is analyzed and obtained. Combined with the landing environment parameter sequence, the control parameters of the air conditioning in the ground area are optimized to obtain the optimal air conditioning control parameters. In this process, temperature weight and energy saving weight are set according to the deviation coefficient between the internal thermal characteristic range of the aircraft and the landing environment parameter sequence for optimization.
[0011] The beneficial effects of this invention are:
[0012] Compared with existing technologies, this invention has the following significant advantages: First, by establishing an environmental parameter prediction mechanism, it can detect environmental changes such as temperature, humidity, and wind speed in the airport area in advance, realizing a shift from passive control to active regulation and significantly improving the system's response accuracy. Second, it innovatively introduces a flight landing time error compensation algorithm, effectively overcoming control deviations caused by flight time uncertainty and avoiding energy waste. Third, by analyzing the internal thermal characteristics of different aircraft models, it establishes a personalized temperature control model, significantly improving passenger comfort. Finally, by employing a multi-objective optimization algorithm, it intelligently balances comfort and energy consumption indicators, significantly reducing the operating energy consumption of the ground air conditioning system while ensuring environmental comfort, thus possessing significant economic benefits and environmental value. Attached Figure Description
[0013] Figure 1 A flowchart illustrating the energy-saving control method for aircraft ground air conditioning that incorporates environmental parameter prediction, provided by this invention.
[0014] Figure 2 A schematic diagram of the energy-saving control method for aircraft ground air conditioning that incorporates environmental parameter prediction provided by the present invention. Detailed Implementation
[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0016] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.
[0017] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.
[0018] Example 1, as Figure 1 , Figure 2 As shown, this embodiment of the invention provides an energy-saving control method for aircraft ground air conditioning based on environmental parameter prediction, including:
[0019] S10: Obtain the aircraft characteristics of the target aircraft, and when the expected landing time of the target aircraft is less than the landing time threshold, obtain the historical environmental parameter sequence of the ground area within the past preset time range;
[0020] The aircraft characteristics of the target aircraft are obtained, and when the expected landing time of the target aircraft is less than a landing time threshold, the historical environmental parameter sequence of the ground area within a preset time range is obtained, including:
[0021] Obtain the aircraft characteristics of the target aircraft, including the aircraft model and service time. The target aircraft is the aircraft to be landed.
[0022] The system obtains the estimated landing time transmitted by the target aircraft. If the landing time is less than a threshold, it obtains the historical environmental parameter sequence of the ground area where the target aircraft will land within a preset time range. Each environmental parameter includes the ambient temperature.
[0023] First, the aircraft characteristics of the target aircraft are acquired, primarily including aircraft model and service time. The target aircraft refers to the aircraft scheduled to land. Second, the estimated landing time transmitted from the target aircraft is obtained, and it is determined whether this time is less than a preset landing time threshold. The landing time threshold refers to the earliest time point that triggers environmental prediction and air conditioning control preparations, set based on airport operational experience and historical flight data statistical analysis, such as 2 or 3 hours in advance. When the condition is met, the data acquisition process is initiated to obtain the historical environmental parameter sequence of the ground area where the target aircraft will land within a preset time range, such as 6 or 8 hours. Environmental parameters include at least ambient temperature. The final obtained historical environmental parameter sequence includes multiple environmental parameter sampling points arranged in chronological order, forming continuous time series data, which facilitates subsequent machine learning models to analyze environmental change trends and make accurate predictions.
[0024] S20: Based on the historical environmental parameter sequence, perform environmental parameter prediction to obtain a predicted environmental parameter sequence within a future preset time range;
[0025] Specifically, based on the historical environmental parameter sequence, environmental parameter prediction is performed to obtain a predicted environmental parameter sequence within a preset future time range, including:
[0026] Acquire environmental parameters to predict the agent;
[0027] The historical environmental parameter sequence is input into the environmental parameter prediction agent, and the prediction output obtains the predicted environmental parameter sequence within a preset time range in the future.
[0028] The environmental parameter prediction agent is a machine learning-based time-series prediction model for environmental parameters. It is obtained through supervised training and can accurately predict the changing trends of environmental parameters in the airport ground area over a future period, providing reliable data support for the forward-looking energy-saving control of aircraft ground air conditioning systems.
[0029] Specifically, the intelligent agent for acquiring environmental parameters includes:
[0030] Based on historical environmental parameter monitoring data of the ground area, a set of historical environmental parameter sequences for samples is collected, and environmental parameters within a preset time range after each historical environmental parameter sequence for samples are collected to obtain a set of predicted environmental parameter sequences for samples.
[0031] Construct an intelligent agent for predicting environmental parameters based on machine learning;
[0032] The environmental parameter prediction agent is trained under supervision using the set of historical environmental parameter sequences and the set of predicted environmental parameter sequences for the samples until the test converges, thus completing the acquisition.
[0033] First, the construction of the environmental parameter prediction agent is based on historical environmental parameter monitoring data from the ground area. It requires collecting a set of historical environmental parameter sequences for each sample sequence, and simultaneously collecting environmental parameter data within a preset time range following each sample sequence, forming a corresponding set of predicted environmental parameter sequences. Specifically, the historical environmental parameter monitoring data refers to time-series data continuously collected by a sensor network deployed in the airport ground area, including multi-dimensional environmental parameters such as temperature, humidity, wind speed, and solar radiation intensity. The environmental parameters within the preset time range following each sample's historical environmental parameter sequence refer to the measured value sequence of the corresponding environmental parameters in subsequent consecutive time periods. The preset time range is a future time span set based on the air conditioning system's regulation characteristics and prediction requirements, such as 1 to 3 hours. The setting of this preset time range needs to comprehensively consider the environmental change patterns and the air conditioning system's response time to ensure that the prediction results not only reflect environmental change trends but also allow sufficient response time for system regulation.
[0034] Secondly, a model architecture for an intelligent agent for predicting environmental parameters is constructed based on machine learning methods. Machine learning is a computer technology that is data-driven, automatically learning from empirical data and improving performance. Constructing an intelligent agent for predicting environmental parameters based on machine learning requires selecting appropriate algorithms, such as Long Short-Term Memory (LSTM) networks, Gated Recurrent Units (GRUs), and Temporal Convolutional Networks (TCNs). In environmental parameter prediction tasks, environmental parameters exhibit significant temporal correlation and periodic variation characteristics; the current environmental state influences future trends, and meteorological patterns evolve dynamically over time, potentially exhibiting seasonal dependencies. For example, the LSTM (Long Short-Term Memory) network model is chosen.
[0035] LSTM (Long Short-Term Memory) is a special type of recurrent neural network designed for processing long-sequence data, particularly adept at capturing long-term dependencies in time series. Its core components include memory cells and gating units. Memory cells act as information channels throughout the network, storing long-term patterns in environmental parameter sequences. As information flows through the cells, the gating units selectively update it: the forget gate determines which historical environmental information needs to be discarded; the input gate determines which new environmental features need to be stored in the memory cell; and the output gate determines which information to output to the next time step based on the current state of the memory cell.
[0036] In practical applications, environmental parameter prediction agents typically employ a multi-layered stacked structure, including: an input layer for receiving time-series environmental feature data, such as temperature, humidity, wind speed, and solar radiation intensity; an LSTM layer as the core computational layer, containing multiple LSTM units that process time-series environmental information through a gating mechanism to capture dependencies between features, such as the correlation between temperature increases and future cooling needs; and a fully connected layer for converting the high-dimensional time-series features output by the LSTM layer into low-dimensional features and integrating key environmental change patterns. The output layer uses a linear activation function to output the predicted environmental parameter sequence within a preset future time range.
[0037] Specifically, the learning rate was set to 0.001, the number of training epochs to 100, and the batch size to 64. The normalized set of historical environmental parameter sequences was input into the environmental parameter prediction agent. The set of predicted environmental parameter sequences was used as the training label. The sample set was divided into a training set (for model learning) and a validation set (for monitoring overfitting) in a 7:3 ratio. During model training, the training set data was used to predict environmental parameters. The difference between the predicted and observed values was calculated using the mean squared error loss function, and the weight parameters of the LSTM and fully connected layers were automatically updated using the backpropagation algorithm. Simultaneously, the model performance was evaluated using the validation set after each training epoch to monitor for overfitting. An early stopping mechanism was used during training; training was automatically terminated when the validation set error no longer decreased for 10 consecutive epochs. Iterative optimization was used to stabilize and reduce the validation set error until a predetermined accuracy requirement was met, such as an average absolute percentage error below 10%. This resulted in an environmental parameter prediction agent capable of accurately predicting changes in environmental parameters. This agent can reliably predict future trends in key parameters such as temperature, humidity, and wind speed based on historical environmental data.
[0038] Finally, the historical environmental parameter sequence is input into the environmental parameter prediction agent, and the prediction output yields the predicted environmental parameter sequence within a preset time range. This predicted environmental parameter sequence accurately reflects future environmental change trends based on historical data, providing crucial data input for the proactive energy-saving control of ground air conditioning systems. It can be used to adjust operating strategies in advance, achieving a balance between energy consumption optimization and precise temperature control.
[0039] S30: Based on the aircraft characteristics, classify the aircraft landing time error, obtain the error coefficient, compensate for the expected landing time to obtain the expected landing time interval, and index the landing environmental parameter sequence within the predicted environmental parameter sequence.
[0040] Based on the aircraft characteristics, the aircraft landing time error is classified to obtain an error coefficient, and the expected landing time is compensated to obtain an expected landing time interval. The landing environment parameter sequence is obtained by indexing within the predicted environment parameter sequence, including:
[0041] The aircraft features are input into the landing time error classification table, and the error coefficients are obtained by classification output. The landing time error classification table is constructed using a sample aircraft feature set and a sample error coefficient set. The error coefficients include the average error magnitude of the aircraft landing time.
[0042] Based on the error coefficient, the estimated landing time is compensated to obtain the estimated landing time range;
[0043] Based on the current time and the expected landing time interval, the expected landing time range is determined, and the predicted environmental parameters within the expected landing time range are indexed in the predicted environmental parameter sequence to obtain the landing environmental parameter sequence.
[0044] First, the acquired aircraft features are input into a pre-defined landing time error classification table. This table is built based on historical flight data and uses regression analysis or machine learning regression algorithms, such as linear regression, gradient boosting regression trees, or support vector regression, to train the model on the deviation between the sample aircraft feature set and the actual landing time, establishing a functional mapping relationship from aircraft features to error coefficients. Specifically, during training, the model learns the quantitative relationship between different aircraft models, service years, airlines, and historical landing time deviations by minimizing the difference between the predicted error and the actual deviation, such as mean squared error. This ultimately forms a regression model that can output continuous error values. The error coefficient is a continuous value obtained through regression analysis, quantitatively representing the average deviation between the landing time and the planned time for this type of aircraft, usually expressed in minutes. For example, an error coefficient of +12 indicates an average delay of 12 minutes for this type of aircraft, while -5 indicates an average arrival time of 5 minutes.
[0045] Secondly, the estimated landing time is compensated and corrected based on the obtained error coefficients. A time window extension method is used, with the estimated landing time as the central reference, extending the time along both the forward and backward time axes by a time length corresponding to one error coefficient, thus forming the estimated landing time interval. This landing time interval represents the range of the aircraft's most likely actual landing time, effectively solving the timing uncertainty problem caused by flight delays or early arrivals.
[0046] Finally, based on the current time and the expected landing time interval, the time range to be queried in the predicted environmental parameter sequence is determined. From the generated predicted environmental parameter sequence, the index extracts the predicted environmental parameters corresponding to all time points within this time range, forming the landing environmental parameter sequence. This landing environmental parameter sequence contains complete predicted environmental parameter values for the aircraft's likely landing period, providing accurate environmental input for subsequent optimization of air conditioning control strategies.
[0047] S40: Based on the aircraft characteristics, analyze and obtain the internal thermal characteristic range of the aircraft. Combined with the landing environment parameter sequence, optimize the control parameters of the air conditioning in the ground area to obtain the optimal air conditioning control parameters. In this process, temperature weight and energy saving weight are set according to the deviation coefficient between the internal thermal characteristic range of the aircraft and the landing environment parameter sequence for optimization.
[0048] Based on the aircraft characteristics, the internal thermal characteristic range of the aircraft is analyzed and obtained. Combined with the landing environment parameter sequence, the control parameters of the ground area air conditioning are optimized to obtain the optimal air conditioning control parameters, including:
[0049] The aircraft features are input into the thermal feature classification table, and the classification output obtains the internal thermal feature range of the aircraft. The thermal feature classification table is constructed using a sample aircraft feature set and a sample internal thermal feature range set. The internal thermal feature range includes the temperature range inside the aircraft.
[0050] Calculate the overlap between the aircraft's internal thermal characteristic range and the landing environment parameter sequence, and obtain the deviation coefficient.
[0051] Based on the ratio of the deviation coefficient to the average deviation coefficient over a historical period, the preset temperature weight is adjusted and calculated to obtain the temperature weight, and the energy-saving weight is calculated.
[0052] Based on the temperature weight and energy-saving weight, the control parameters of the ground area air conditioning are optimized to obtain the optimal air conditioning control parameters.
[0053] First, aircraft characteristics are input into a pre-constructed thermal feature classification table. This table employs a supervised learning method, establishing a feature mapping relationship by training a set of sample aircraft features, such as aircraft type, service life, and number of seats, with the corresponding set of internal thermal feature intervals. The internal thermal feature interval refers to the typical temperature distribution range inside the cabin of an aircraft of that model under specific seasons and operating conditions, for example, [23℃, 26℃]. This interval is obtained through statistical analysis of historical cabin temperature monitoring data. The thermal feature classification table establishes a non-linear mapping relationship from aircraft type, service life, and number of seats to cabin thermal feature intervals, enabling rapid and accurate inference of the internal thermal environment requirements of the target aircraft based on its individual characteristics, providing crucial input for subsequent personalized optimization of air conditioning control parameters.
[0054] Secondly, the percentage of overlap between the aircraft's internal thermal characteristic range and the landing environmental parameter sequence is calculated. This percentage represents the proportion of the predicted environmental temperature falling within the cabin's comfortable temperature range, calculated as: Percentage of overlap = (Number of data points in the overlap range) / (Total number of data points in the environmental sequence). The deviation coefficient is defined as: Deviation coefficient = 1 - Percentage of overlap. This deviation coefficient quantitatively characterizes the degree of deviation between environmental conditions and the aircraft's internal thermal requirements; a larger deviation coefficient indicates a more unsuitable environment and a greater need for temperature control.
[0055] For example, suppose the predicted ambient temperature sequence is [28℃, 27℃, 26℃, 25℃, 27℃], and the internal thermal characteristic range of a certain type of aircraft is [23℃, 26℃]. Then the intersection ratio = 2 / 5 = 0.4, and the deviation coefficient = 1 - 0.4 = 0.6.
[0056] Furthermore, based on this deviation coefficient, optimization weights are dynamically configured: the larger the deviation coefficient, the higher the value of the temperature weight is set, and the energy-saving weight is correspondingly reduced, thereby prioritizing temperature control accuracy and user experience, ensuring that the cabin quickly reaches and maintains a comfortable temperature; the smaller the deviation coefficient, the higher the energy-saving weight is set, and the temperature weight is correspondingly reduced, prioritizing energy efficiency optimization when environmental conditions are relatively suitable, allowing temperature fluctuations within a small range to achieve energy-saving goals. For example, the temperature weight Wt = deviation coefficient, and the energy-saving weight We = 1 - Wt.
[0057] For example, when the deviation coefficient is 0.6, the temperature weight Wt = deviation coefficient = 0.6, and the energy saving weight We = 1 - Wt = 0.4.
[0058] Furthermore, based on the temperature weight and energy-saving weight, the control parameters of the ground area air conditioning are optimized to obtain the optimal air conditioning control parameters, including:
[0059] Obtain the control parameter range of the air conditioning in the ground area, and randomly generate the first air conditioning control parameter;
[0060] Based on the first air conditioning control parameters and the landing environment parameter sequence, the first control temperature is predicted and the first temperature adaptability is calculated.
[0061] Obtain the first air conditioner energy consumption parameter from the first air conditioner control parameters, calculate the ratio of the standard air conditioner energy consumption parameter to the first air conditioner energy consumption parameter, and obtain the first energy-saving adaptability.
[0062] Based on the temperature weight and energy-saving weight, the first temperature adaptability and the first energy-saving adaptability are weighted and calculated to obtain the first air conditioning adaptability;
[0063] Continue iteratively optimizing the air conditioning control parameters until convergence, obtaining the optimal air conditioning control parameters with the greatest air conditioning adaptability.
[0064] First, obtain the adjustable control parameter ranges for the ground area air conditioning, such as compressor frequency range, fan speed range, and refrigerant flow rate range. Within this adjustable control parameter range, randomly generate a set of initial parameter combinations, denoted as the first air conditioning control parameter, as the starting point for the optimization algorithm.
[0065] Secondly, based on the first air conditioning control parameters and the landing environment parameter sequence, a first control temperature is predicted and a first temperature adaptability is calculated, including:
[0066] Calculate the mean of the landing environment parameter sequence to obtain the average landing environment parameters;
[0067] The first air conditioning control parameter and the average landing environment parameter are input into the air conditioning control prediction table, and the first control temperature is output. The air conditioning control prediction table is constructed using a sample air conditioning control parameter set, a sample environment parameter set, and a sample control temperature set.
[0068] Calculate the similarity between the first control temperature and the standard temperature to obtain the first temperature adaptability.
[0069] First, the arithmetic mean of the predicted landing environment parameter sequence is calculated to obtain the average landing environment parameter. This operation simplifies the dynamic environment sequence into a representative static parameter, which characterizes the overall environmental conditions during the aircraft landing period, thereby reducing the complexity of subsequent prediction models. Second, the first air conditioning control parameter and the calculated average landing environment parameter are input into a pre-built air conditioning control prediction table. This air conditioning control prediction table is a machine learning model built based on historical operating data. Its core function is to establish a mapping relationship from control parameters and environmental parameters to cabin temperature. For example, due to the highly nonlinear and complex correlation between air conditioning control parameters, environmental parameters, and cabin temperature, and the significant advantages of neural networks in automatic feature extraction and fitting of complex nonlinear relationships, a neural network model is chosen to construct the air conditioning control prediction table.
[0070] Specifically, the air conditioning control prediction model mainly consists of an input preprocessing layer, a feature extraction layer, and an output layer. The input preprocessing layer receives raw data such as compressor frequency, fan speed, and average environmental parameters, and eliminates dimensional differences through normalization. The feature extraction layer employs a multi-layer fully connected neural network structure, with the number of neurons adaptively configured according to the feature dimensions. Each layer is equipped with a ReLU activation function to introduce nonlinear transformation capabilities, and Dropout layers (with a dropout rate set between 0.2 and 0.5) are embedded between layers to suppress overfitting and improve the model's generalization performance. The output layer uses a single neuron equipped with a linear activation function to map the final features to continuous predicted temperature values, outputting the first control temperature.
[0071] Secondly, the air conditioning control prediction model is trained using supervised learning. Specifically, sample sets of air conditioning control parameters and corresponding sample sets of environmental parameters are collected from historical operating data of the air conditioning system as input sample sets; simultaneously, the actual cabin temperature after stabilization under the corresponding parameters is acquired to form a sample set of control temperatures as a label sample set. The input sample set and the corresponding label sample set are divided into training set, validation set, and test set in a 7:2:1 ratio. Then, the sample parameters in the training set are used as input features, and the corresponding sample control temperatures are used as supervision labels. The network weight parameters are iteratively optimized using the backpropagation algorithm and the Adam optimizer. The mean squared error loss function is used to measure the deviation between the predicted temperature and the actual temperature, and the training process is monitored through the validation set. When the validation set loss no longer decreases for several consecutive rounds (e.g., 10 rounds) and the mean absolute error stably drops to a predetermined threshold (e.g., ±0.5℃), the training is terminated, and a converged air conditioning control prediction model is obtained. This model is used to capture the complex nonlinear relationship between control parameters, environmental parameters, and cabin temperature to achieve high-precision temperature prediction.
[0072] Finally, the real-time acquired air conditioning control parameters and average landing environment parameters are input into the trained air conditioning control prediction table. Based on the learned system characteristic mapping relationship, calculations are performed to output the first control temperature. This first control temperature is used for subsequent calculations of temperature fitness, providing a key evaluation indicator for optimizing air conditioning control parameters.
[0073] Furthermore, the predicted first control temperature is compared with the standard temperature, i.e., the desired target cabin temperature, which is usually the median of the comfort range, such as 24°C. The similarity between the two is calculated as the first temperature fitness. First temperature fitness = 1 / (1 + |first control temperature - standard temperature|). The smaller the temperature difference, the larger the fitness value, indicating a better control effect; the larger the temperature difference, the smaller the fitness value, indicating a worse control effect.
[0074] For example, if the predicted first control temperature is 25.5℃ and the standard temperature is 24℃, then the first temperature adaptability = 1 / [1+(25.5-24)] = 0.4.
[0075] Simultaneously, the corresponding first air conditioner energy consumption parameters are queried or calculated based on the first air conditioner control parameters. The first energy-saving adaptability is calculated as the ratio of the standard air conditioner energy consumption parameters, such as the energy consumption under rated operating conditions or the historical average energy consumption, to the first air conditioner energy consumption parameters. First energy-saving adaptability = Standard air conditioner energy consumption parameters / First air conditioner energy consumption parameters. If the first energy-saving adaptability ratio is greater than 1, it indicates that the actual energy consumption is lower than the standard energy consumption benchmark, the higher the energy efficiency, and the better the energy-saving effect; conversely, if the ratio is less than 1, it indicates that the energy efficiency is low and the energy-saving effect is poor.
[0076] For example, if the estimated first air conditioner energy consumption parameter is 42kWh and the standard air conditioner energy consumption parameter is 50kWh, then the first energy-saving adaptability = 50 / 42≈1.19, which shows excellent energy-saving effect and the actual energy consumption is lower than the standard energy consumption.
[0077] Furthermore, based on the previously determined temperature weight Wt and energy-saving weight We, the first temperature fitness and the first energy-saving fitness are weighted and summed to obtain the first air conditioning fitness, which is calculated as: First Air Conditioning Fitness = Wt × First Temperature Fitness + We × First Energy-Saving Fitness. This fitness is used to comprehensively evaluate the overall performance of the current air conditioning control parameters in terms of temperature control effect and energy consumption efficiency, providing a single, quantitative optimization objective for the optimization algorithm and guiding the search direction to find the optimal solution that can simultaneously balance comfort and economy.
[0078] For example, if the temperature weight Wt is 0.6, the energy-saving weight We is 0.4, the first temperature adaptability is 0.4, and the first energy-saving adaptability is 1.19, then the first air conditioning adaptability = (0.6 * 0.4) + (0.4 * 1.19) = 0.716. The advantage of energy saving is not enough to completely make up for the lack of comfort, and we need to continue to look for a better solution.
[0079] Finally, the air conditioning control parameters are iteratively optimized until convergence, yielding the optimal air conditioning control parameters with the highest fitness. Specifically, a population-based optimization algorithm, such as particle swarm optimization or genetic algorithm, is used, employing air conditioning fitness as the evaluation function to perform a global search within the safe operating range of the air conditioning control parameters. In each iteration, the algorithm generates new parameter combinations and quickly evaluates their corresponding temperature fitness and energy-saving fitness using the constructed air conditioning control prediction table and energy consumption calculation model, thereby calculating the comprehensive air conditioning fitness. Furthermore, high-performance parameter combinations are retained based on their fitness values, while low-performance combinations are eliminated. New solution spaces are explored through mechanisms such as crossover, mutation, or particle position updates. When, after multiple iterations (e.g., 10 generations), the improvement in the optimal fitness in the population is less than a preset threshold (e.g., 0.001), or when the maximum number of iterations is reached (e.g., 100 generations), the algorithm is considered converged. At this point, the parameter combination with the highest air conditioning fitness obtained during the global search is the optimal air conditioning control parameter, which achieves the optimal balance between comfort and energy efficiency under specific environmental conditions.
[0080] In summary, the embodiments of this application have at least the following technical effects:
[0081] Compared to existing technologies, this invention firstly introduces an environmental parameter prediction mechanism, enabling it to anticipate changes in environmental factors such as temperature, humidity, and wind speed in airport areas, thus shifting from passive response to proactive control and significantly improving the foresight and accuracy of the control. Secondly, by combining aircraft characteristics for landing time error classification and compensation, it effectively overcomes the time uncertainty caused by flight delays or early arrivals, ensuring that the air conditioning system starts at the optimal time and matches the environmental conditions at the actual landing time, avoiding ineffective energy consumption. Thirdly, by analyzing the internal thermal characteristic ranges of different aircraft models and dynamically optimizing the control strategy based on their deviation from predicted environmental parameters, it achieves personalized temperature adjustment for different aircraft, improving passenger comfort while avoiding excessive cooling or heating. Finally, by setting temperature weights and energy-saving weights for multi-objective optimization, it can intelligently balance energy consumption and effectiveness while ensuring comfort, achieving refined energy-saving control and effectively reducing the overall operating cost of airport ground air conditioning systems, demonstrating strong practicality and promotional value.
[0082] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.
[0083] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.
[0084] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.
Claims
1. An energy-saving control method for aircraft ground air conditioning based on environmental parameter prediction, characterized in that, The method includes: The aircraft characteristics of the target aircraft are obtained, and when the expected landing time of the target aircraft is less than a landing time threshold, the historical environmental parameter sequence of the ground area within a preset time range is obtained, including: Obtain the aircraft characteristics of the target aircraft, including the aircraft model and service time. The target aircraft is the aircraft to be landed. The system obtains the estimated landing time transmitted by the target aircraft. If the landing time is less than the landing time threshold, it obtains the historical environmental parameter sequence of the ground area where the target aircraft will land within a preset time range. Each environmental parameter includes the ambient temperature. Based on the historical environmental parameter sequence, environmental parameter prediction is performed to obtain a predicted environmental parameter sequence within a preset time range in the future; Based on the aircraft characteristics, the aircraft landing time error is classified to obtain the error coefficient, and the expected landing time is compensated to obtain the expected landing time interval. The landing environmental parameter sequence is obtained by indexing within the predicted environmental parameter sequence. Based on the aircraft characteristics, the internal thermal characteristic range of the aircraft is analyzed and obtained. Combined with the landing environment parameter sequence, the control parameters of the ground area air conditioning are optimized to obtain the optimal air conditioning control parameters. This optimization involves setting temperature weights and energy-saving weights based on the deviation coefficient between the internal thermal characteristic range of the aircraft and the landing environment parameter sequence, including: The aircraft features are input into the thermal feature classification table, and the classification output obtains the internal thermal feature range of the aircraft. The thermal feature classification table is constructed using a sample aircraft feature set and a sample internal thermal feature range set. The internal thermal feature range includes the temperature range inside the aircraft. Calculate the overlap between the aircraft's internal thermal characteristic range and the landing environment parameter sequence, and obtain the deviation coefficient. Based on the ratio of the deviation coefficient to the average deviation coefficient over a historical period, the preset temperature weight is adjusted and calculated to obtain the temperature weight, and the energy-saving weight is calculated. Based on the temperature weight and energy-saving weight, the control parameters of the ground area air conditioning are optimized to obtain the optimal air conditioning control parameters.
2. The aircraft based on environmental parameter prediction as described in claim 1 The energy-saving control method for ground air conditioning is characterized by, Based on the historical environmental parameter sequence, environmental parameter prediction is performed to obtain a predicted environmental parameter sequence within a preset future time range, including: Acquire environmental parameters to predict the agent; The historical environmental parameter sequence is input into the environmental parameter prediction agent, and the prediction output obtains the predicted environmental parameter sequence within a preset time range in the future.
3. The aircraft based on environmental parameter prediction as described in claim 2 The energy-saving control method for ground air conditioning is characterized by, The agent obtains environmental parameters for prediction, including: Based on historical environmental parameter monitoring data of the ground area, a set of historical environmental parameter sequences for samples is collected, and environmental parameters within a preset time range after each historical environmental parameter sequence for samples are collected to obtain a set of predicted environmental parameter sequences for samples. Construct an intelligent agent for predicting environmental parameters based on machine learning; The environmental parameter prediction agent is trained under supervision using the set of historical environmental parameter sequences and the set of predicted environmental parameter sequences for the samples until the test converges, thus completing the acquisition.
4. The aircraft based on environmental parameter prediction as described in claim 1 The energy-saving control method for ground air conditioning is characterized by, Based on the aircraft characteristics, the aircraft landing time error is classified to obtain an error coefficient, and the expected landing time is compensated to obtain an expected landing time interval. The landing environment parameter sequence is obtained by indexing within the predicted environment parameter sequence, including: The aircraft features are input into the landing time error classification table, and the error coefficients are obtained by classification output. The landing time error classification table is constructed using a sample aircraft feature set and a sample error coefficient set. The error coefficients include the average error magnitude of the aircraft landing time. Based on the error coefficient, the estimated landing time is compensated to obtain the estimated landing time range; Based on the current time and the expected landing time interval, the expected landing time range is determined, and the predicted environmental parameters within the expected landing time range are indexed in the predicted environmental parameter sequence to obtain the landing environmental parameter sequence.
5. The aircraft based on environmental parameter prediction as described in claim 1 The energy-saving control method for ground air conditioning is characterized by, Based on the temperature weight and energy-saving weight, the control parameters of the ground area air conditioning are optimized to obtain the optimal air conditioning control parameters, including: Obtain the control parameter range of the air conditioning in the ground area, and randomly generate the first air conditioning control parameter; Based on the first air conditioning control parameters and the landing environment parameter sequence, the first control temperature is predicted and the first temperature adaptability is calculated. Obtain the first air conditioner energy consumption parameter from the first air conditioner control parameters, calculate the ratio of the standard air conditioner energy consumption parameter to the first air conditioner energy consumption parameter, and obtain the first energy-saving adaptability. Based on the temperature weight and energy-saving weight, the first temperature adaptability and the first energy-saving adaptability are weighted and calculated to obtain the first air conditioning adaptability; Continue iteratively optimizing the air conditioning control parameters until convergence, obtaining the optimal air conditioning control parameters with the greatest air conditioning adaptability.
6. The aircraft based on environmental parameter prediction as described in claim 5 The energy-saving control method for ground air conditioning is characterized by, Based on the first air conditioning control parameters and the landing environment parameter sequence, a first control temperature is predicted, and a first temperature adaptability is calculated, including: Calculate the mean of the landing environment parameter sequence to obtain the average landing environment parameters; The first air conditioning control parameter and the average landing environment parameter are input into the air conditioning control prediction table, and the first control temperature is output. The air conditioning control prediction table is constructed using a sample air conditioning control parameter set, a sample environment parameter set, and a sample control temperature set. Calculate the similarity between the first control temperature and the standard temperature to obtain the first temperature adaptability.