Railway locomotive cab temperature self-adaptive adjusting method

Abstract

The invention relates to the technical field of temperature regulation and control, in particular to a railway locomotive cab temperature self-adaptive regulation method which comprises the following steps: S1, dividing a target area into a plurality of target areas, and sequentially acquiring a plurality of historical data samples of each target area under each target solar term, the historical data sample comprises driver indoor and outdoor historical temperature, locomotive running speed, solar term labels and latitude and longitude information, and obtaining an actual data sample of a current target area in a current target solar term through an information acquisition module; the temperature change direction and amplitude can be pre-judged in advance, air conditioner power adjustment or ventilation mode switching can be started in advance according to the temperature change direction and amplitude, sudden rise and sudden drop caused by passive adjustment after the temperature breaks through a comfort threshold value are avoided, temperature control logic of active pre-judgment and pre-adjustment is constructed, temperature fluctuation of a cab is effectively reduced, the driving comfort of a driver is remarkably improved, and the driving safety of the driver is improved. Therefore, the attention of the driver during long-time driving is concentrated, and the driving safety is indirectly assisted.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of temperature regulation, in particular to a railway locomotive cab temperature self-adaptive regulation method. BACKGROUND

[0002] The railway locomotive cab is equipped with an air conditioner as an auxiliary device for improving the indoor environment temperature. With the change of the vehicle running position, the altitude and the environment temperature will change, and the indoor environment temperature of the cab will change. The driver will feel overheating or overcooling. To effectively regulate the indoor environment temperature of the cab, the air conditioner must be frequently started and stopped. The traditional regulation method is divided into two types. One is manual regulation, which requires the driver to frequently operate according to subjective feeling. This not only disperses the driving attention and increases the safety risk, but also the regulation accuracy is greatly affected by the driver's experience. The other is fixed temperature threshold triggering. The main process is that the real-time temperature acquisition module collects the temperature, and the temperature controller judges whether the set value of the temperature setting module is reached. If not, the temperature controller controls the air conditioner to run. The fixed threshold regulation does not consider the seasonal and regional differences. For example, the outdoor baseline temperature difference between the northern winter (-20℃~0℃) and the southern winter (5℃~15℃) is significant. Under the same threshold, the problem of insufficient heating in the north and excessive heating in the south is easy to appear. It cannot match the heating demand in different environments. Moreover, the above two traditional regulation methods are "passive response". The air conditioner is started and stopped according to the real-time collected cab temperature, which belongs to the passive regulation mode of temperature intervention after exceeding the standard. In the railway locomotive operation, the environment temperature is dynamically changed by factors such as regional switching, day and night alternation, and tunnel entry and exit. Passive regulation is easy to cause large fluctuation of the cab temperature, which affects the comfort of the driver and increases the equipment wear and energy waste due to frequent start and stop of the air conditioner. SUMMARY

[0003] In view of the defects of the prior art, the present application provides a railway locomotive cab temperature self-adaptive regulation method, which solves the technical problems of low regulation accuracy, large temperature fluctuation, difficulty in adapting to complex scenes across regions and seasons, high cost of manual intervention, and insufficient energy consumption optimization caused by dependence on a single parameter and passive response in traditional temperature regulation.

[0004] To solve the above technical problems, the present application provides the following technical scheme: a railway locomotive cab temperature self-adaptive regulation method, which comprises the following steps: S1. Divide the target area into several target regions, and sequentially obtain several historical data samples of each target region under each target solar term. The historical data samples include historical indoor and outdoor temperatures of the driver's cab, locomotive operating speed, solar term labels, and latitude and longitude information. Obtain actual data samples of the current target region under the current target solar term through the information acquisition module. The actual data samples include actual indoor and outdoor temperatures of the driver's cab, actual locomotive operating speed, solar term labels, and latitude and longitude information. S2. Construct a temperature prediction model that integrates the temperature regulation strategy based on the solar terms and the temperature regulation strategy based on the region, and optimize the model parameters through iterative training using historical data samples to obtain the trained temperature prediction model. S3. Input the actual data sample into the trained temperature prediction model and output the predicted temperature change trend of the driver's cab within a preset time period. S4. Compare the predicted temperature change trend with the preset target temperature range for the driver's cab to determine whether the future temperature deviates from the target temperature range. If so, proceed to step S5; If not, the controller will not generate an adjustment command; S5. The controller generates the corresponding temperature adjustment command and sends it to the temperature actuator. The temperature actuator receives the temperature adjustment command and executes the corresponding temperature adjustment action to maintain the temperature in the driver's cab within the target temperature range.

[0005] Preferably, in step S2, the step of constructing the temperature prediction model is as follows: S21. Clean the historical data samples to generate a standard dataset as training samples. S22. Construct a temperature prediction sub-model based on solar terms using standard datasets. and region-based temperature prediction sub-model ; S23, The temperature prediction sub-model based on solar terms and region-based temperature prediction sub-model By fusion, a temperature prediction model is obtained. Its expression is: ; In the above formula, The dynamic weight of the solar term at time t. Let be the regional dynamic weight at time t; S24. Temperature prediction model using standard dataset To perform training and obtain the trained model parameters and The trained temperature prediction model is obtained. The objective function for model training is: ; In the above formula, Let T be the baseline value for the weight of the solar term. Let T be the baseline value for the regional weight. The actual temperature values ​​at time t during the S solar term and along route L. To optimize the objective function that compares the differences between the dynamic weights of solar terms and regional dynamic weights and the benchmark weights, The objective function is the difference between the model-predicted temperature and the baseline-predicted temperature.

[0006] Preferably, in step S22, the construction steps of the temperature prediction sub-model based on the solar terms are as follows: S2211. Calculate the reference temperature for a given solar term using the reference value of the driver's cab temperature under a standard scenario within that solar term. The calculation formula is: ; In the above formula, This is the line-level reference temperature correction factor for route L during throttle S. For vehicle model correction; S2212. Calculate the temperature change rate of a solar term based on the average rate of temperature change over time within that solar term. The calculation formula is: ; In the above formula, This represents the average rate of change within the same solar term and time period group. This is the solar radiation correction factor. This is the vehicle speed correction factor; S2213, Based on the throttle reference temperature and the rate of change of temperature during the solar terms Construct a temperature prediction sub-model based on solar terms The expression is: ; In the above formula, For any time in the future With the current moment The time difference.

[0007] Preferably, in step S2211, the calculation steps for the line-level reference temperature correction coefficient are as follows: S22111. Calculate the average temperature in the driver's cab during a certain throttle period based on a standard dataset. The calculation formula is: ; In the above formula, This represents the total number of valid data entries within solar term S. This is the i-th valid temperature data point within solar term S; S22112, Calculation of Line Climate Correction Coefficient Based on Line Outdoor Temperature The calculation formula is: ; In the above formula, Let L be the average outdoor temperature of line L during the solar term S. The average outdoor temperature for all lines during solstice S; S22113, The average temperature of the driver's cab during a certain solar term. With line climate correction factor The line-level reference temperature correction factor was obtained by calculation. The calculation formula is: .

[0008] Preferably, in step S2212, the calculation steps for the average change rate within the same solar term and time period group are as follows: S22121. Filter the standard dataset to obtain continuous temperature sequences, and use the least squares method to calculate the original rate of change of a single sequence within the same solar term and time period group. The calculation formula is: ; In the above formula, The number of data points in the sequence. Relative time within the sequence, The relative temperature within the sequence; S22122. For all the original rates of change in the same solar term and time period group, after removing the 10% extreme values, the average rate of change within the group is obtained through smoothing. The calculation formula is: ; In the above formula, This represents the total number of original rate of change entries within the group. To eliminate the 10% extreme values, This represents the j-th original rate of change after sorting.

[0009] Preferably, in step S22, the construction steps of the region-based temperature prediction sub-model are as follows: S2221. Calculate the regional environmental reference temperature using the driver's cab temperature reference value under a standard scenario within a certain region. The calculation formula is: ; In the above formula, The baseline average temperature for route L. The continentality index, The average continentality index for all regions, with a coefficient of 0.005 representing the correction ratio for every 1 unit deviation in continentality; S2222. Calculate the rate of temperature change in a region by using the average rate of temperature change over time within that region. The calculation formula is: ; In the above formula, Let L be the average rate of change within the regional group of route L. This is a correction factor for outdoor temperature fluctuations. This is a regional vehicle speed correction factor; S2223, Based on regional environmental reference temperature and regional temperature change rate Constructing a region-based temperature prediction sub-model Its expression is: ; In the above formula, For any time in the future With the current moment The time difference.

[0010] Preferably, in step S2221, the specific implementation steps are as follows: S22211. Calculate the average temperature of the driver's cab in a certain region based on a standard dataset. The calculation formula is: ; In the above formula, This represents the total number of valid data entries within region L. This is the i-th valid temperature data point within region L; S22212, Average value corrected by altitude To obtain the base average temperature The calculation formula is: ; In the above formula, This represents the actual altitude of the current location. For reference altitude, a coefficient of 0.006 represents the temperature correction per meter of altitude. S22213, Calculate the continentality index The calculation formula is: ; In the above formula, This refers to the annual temperature range in this region; S22214, Based on the average temperature And Continental Index The regional environmental reference temperature was calculated. .

[0011] Preferably, in step S2222, the calculation steps for the average rate of change within the regional group are as follows: S22221. Filter the standard dataset to obtain continuous temperature sequences, and use the least squares method to calculate the original rate of change of a single sequence within the same region and time period group. The calculation formula is: ; In the above formula, The number of data points in the sequence. Relative time within the sequence, The relative temperature within the sequence; S22222. For all the original rates of change in the same region and time period group, remove the 10% extreme values ​​and then smooth them to obtain the average rate of change within the region group. The calculation formula is: ; In the above formula, This represents the total number of original rate of change entries within the group. To eliminate the 10% extreme values, This represents the j-th original rate of change after sorting.

[0012] Preferably, in step S23, the calculation steps for the solar term dynamic weight and the regional dynamic weight are as follows: S231. Extracting solar terms from the calendar Calculate the solar term intensity factor by combining the periodic characteristics of the solar terms. The calculation formula is: ; In the above formula, The 24 solar terms The time difference between the current date and the S-th solar term. The cosine function of the intensity decay of the solar term. The basic importance weight of the Sth solar term; S232. Derive the continentality index using latitude and longitude information, and then use the normalized continentality index... and altitude correction factor Comprehensive calculation of regional intensity factor The calculation formula is: ; In the above formula, The normalized continentality index. Altitude correction factor; S233, Regarding the intensity factor of solar terms and regional intensity factor Normalization is performed to generate initial weights, and the expressions for both are as follows: ; In the above formula, The interaction coefficient; S234. Combining real-time prediction errors and regional change rates, optimize the initial weights to obtain dynamic weights for each solar term. and regional dynamic weights The expressions for the two are as follows: ; In the above formula, The function's weights are restricted to a range of 0.1-0.9. To adjust the rate, It is the minimum value. Temperature prediction sub-model based on solar terms The prediction error Location-based temperature prediction submodel The prediction error.

[0013] By employing the above technical solution, the present invention provides a method for adaptive temperature regulation in a railway locomotive driver's cab, which has at least the following beneficial effects: 1. This invention integrates a dual temperature regulation strategy based on solar terms and regional factors, and trains the model using historical data samples from railway locomotive operations. On the one hand, the solar term strategy anchors the temperature benchmark for different solar terms, while the regional strategy supplements local climate differences, thus covering core environmental variables in two dimensions. On the other hand, historical data training allows the model to learn the temperature change patterns of specific lines and locomotive models, enabling temperature prediction and regulation to accurately match actual operating scenarios. This avoids regulation deviations caused by the omission of environmental variables, ensuring the accuracy of temperature control for different lines and locomotive models under diverse climates. It fundamentally solves the drawbacks of traditional single-parameter regulation that is "one-size-fits-all," and achieves precise adaptation of temperature regulation to specific scenarios.

[0014] 2. This invention outputs the temperature change trend in real time for future periods through a trained temperature prediction model. When the locomotive is about to enter a sun-exposed road section, a high-altitude low-temperature area, or a long tunnel, the model can predict the direction and magnitude of temperature changes in advance and start adjusting the air conditioning power or switching the ventilation mode accordingly. This avoids sudden rises and falls caused by passive adjustment after the temperature exceeds the comfort threshold. It constructs a temperature control logic that actively predicts and adjusts in advance, effectively reducing temperature fluctuations in the driver's cab. It can control the temperature fluctuation range in the driver's cab within ±1.5°C, significantly improving the driver's driving comfort, thereby ensuring the driver's concentration during long-distance driving and indirectly contributing to driving safety.

[0015] 3. The temperature prediction model of this invention can autonomously adapt to changes in multiple scenarios. For example, when a locomotive is running across regions, the model can automatically adjust the regional strategy weights based on latitude and longitude to adapt to the characteristics of high altitude, low temperature, and large temperature difference between day and night. When the solar terms change, the temperature prediction model can automatically update the solar term benchmark temperature and change rate through solar term and month parameters without manual intervention. This can significantly reduce manual operation and maintenance costs, reduce human error, and make the temperature control system more in line with the long-distance and multi-scenario operation needs of railway locomotives.

[0016] 4. This invention can reduce the ineffective energy consumption of temperature control equipment by 15%-20%, while reducing the frequent switching between high load and shutdown states of the equipment, reducing the failure probability of core components such as compressors and fans, extending the maintenance cycle of temperature control equipment, saving railway operating units equipment procurement and maintenance costs, and achieving the dual benefits of energy saving and cost reduction. Attached Figure Description

[0017] The accompanying drawings, which are provided to further illustrate the invention, constitute a part of this application: Figure 1 This is a flowchart of the adaptive temperature control method for railway locomotive driver's cab according to the present invention. Detailed Implementation

[0018] 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.

[0019] Given the limitations of existing technologies—such as low adjustment precision and large temperature fluctuations due to reliance on a single parameter and passive response, difficulty in adapting to complex scenarios across regions and seasons, high cost of manual intervention, and insufficient energy consumption optimization—please refer to [the relevant documentation / reference]. Figure 1 This embodiment provides a method for adaptive temperature adjustment in a railway locomotive driver's cab. It can predict the direction and magnitude of temperature changes in advance and accordingly initiate adjustments to the air conditioning power or switch ventilation modes. This avoids sudden increases or decreases caused by passive adjustments after the temperature exceeds the comfort threshold. It establishes a proactive temperature control logic from prediction to advance adjustment, effectively reducing temperature fluctuations in the driver's cab, significantly improving driver comfort, and thus ensuring the driver's concentration during long driving periods, indirectly contributing to driving safety. The adjustment method includes the following steps: S1. Divide the target region into several target areas and obtain several historical data samples of each target area under each target solar term. The historical data samples include the historical temperature inside and outside the driver's cab, the locomotive's operating speed, the solar term label, and latitude and longitude information. Obtain the actual data sample of the current target area under the current target solar term through the information collection module. The actual data sample includes the actual temperature inside and outside the driver's cab, the actual locomotive's operating speed, the solar term label, and latitude and longitude information. Taking South China as the target region, it is divided into three target areas according to topography and climate: the Pearl River Delta Plain (A), the Nanling Mountains (B), and the Hainan Coastal Area (C). Select the 24 solar terms as the target solar terms. For example, historical data samples collected in Region A (Pearl River Delta Plain) during the summer solstice period from 2022 to 2024 include: on June 22, 2023, the temperature in the driver's cab of a freight train on the Guangzhou-Dongguan section was 28.5℃, the outdoor temperature was 34.2℃, the average speed was 60km / h, the solar term label was "Summer Solstice", and the latitude and longitude were 22.5°N-22.8°N and 113.8°E-113.9°E, as well as more than 500 other similar data covering different train types and time periods. On June 22, 2025 (summer solstice), when a locomotive was traveling on the Shenzhen-Foshan section of Region A, the real-time data samples obtained by the information collection module were: the temperature in the driver's cab was 29.1℃, the outdoor temperature was 35.0℃, the real-time speed was 200km / h, the solar term label was "Summer Solstice", and the latitude and longitude were 22.6°N and 113.7°E. This historical data was used to train model parameters, while the actual data was input into the model to predict future temperature changes. S2. Construct a temperature prediction model that integrates solar term-based and region-based temperature regulation strategies. Iteratively train and optimize the model parameters using historical data samples to obtain a trained temperature prediction model. This model outputs real-time temperature change trends for future time periods, avoiding sudden rises and falls caused by passive adjustments after temperatures exceed comfort thresholds. This constructs a proactive temperature control logic that predicts and adjusts in advance, effectively reducing temperature fluctuations in the driver's cab, significantly improving driver comfort, and thus ensuring the driver's concentration during long drives, indirectly contributing to driving safety. The steps for constructing the temperature prediction model are as follows: S21. Perform data cleaning on historical data samples to generate a standard dataset for training. Data cleaning includes removing outliers, removing invalid scenarios, and imputing missing values. Among these, outlier removal uses the 3σ principle to filter out values ​​exceeding the driver's cab temperature limit. Data ( This is the historical average. (Standard deviation) Remove invalid scenarios by removing data from extreme weather (rainstorm, high temperature warning) and special operating conditions (vehicle speed <10km / h, staying in tunnel). Fill missing values ​​are filled by linear interpolation for short-term missing values ​​and the sequence of the time period is directly removed for long-term missing values. S22. Construct a temperature prediction sub-model based on solar terms using standard datasets. and region-based temperature prediction sub-model ; When the solar terms change, the temperature prediction model can automatically update the solar term baseline temperature and rate of change based on the solar term and month parameters. The construction steps of the solar term-based temperature prediction sub-model are as follows: S2211. Calculate the reference temperature for a given solar term using the reference value of the driver's cab temperature under a standard scenario within that solar term. The standard scenario specifically refers to a scenario with normal vehicle speed and no extreme weather conditions. The calculation formula is as follows: ; In the above formula, This is the line-level reference temperature correction factor for route L during throttle S. The model correction amount can be determined through comparative testing, such as +3℃ for diesel locomotives, 0℃ for electric locomotives, and -2℃ for EMUs. The calculation steps for the line-level reference temperature correction coefficient are as follows: S22111. Calculate the average temperature in the driver's cab during a certain throttle period based on a standard dataset. The calculation formula is: ; In the above formula, This represents the total number of valid data entries within solar term S. This is the i-th valid temperature data point within solar term S; S22112, Calculation of Line Climate Correction Coefficient Based on Line Outdoor Temperature The calculation formula is: ; In the above formula, The average outdoor temperature of line L during solar term S is obtained directly from the meteorological database. The constant 5 represents the average outdoor temperature of all lines during the solar term S, which helps to prevent excessive fluctuations in the coefficient when the outdoor temperature is close to 0°. S22113, The average temperature of the driver's cab during a certain solar term. With line climate correction factor The line-level reference temperature correction factor was obtained by calculation. The calculation formula is: .

[0020] S2212. Calculate the temperature change rate of a solar term based on the average rate of temperature change over time within that solar term. This reflects how much the temperature rises or falls per hour, and the calculation formula is: ; In the above formula, This represents the average rate of change within the same solar term and time period group. This is a solar radiation correction factor; temperatures rise rapidly during the day due to sunlight. The speed correction factor is used; the higher the speed, the stronger the ventilation, and the more pronounced the change. The calculation steps for the intra-group average rate of change within the same solar term and time period are as follows: S22121. Filter the standard dataset to obtain continuous temperature sequences, and use the least squares method to calculate the original rate of change of a single sequence within the same solar term and time period group. The calculation formula is: ; In the above formula, This refers to the number of data points in the sequence; for example, a 2-hour sequence contains 24 data points. Relative time within the sequence, The relative temperature within the sequence; S22122. For all the original rates of change in the same solar term and time period group, after removing the 10% extreme values, the average rate of change within the group is obtained through smoothing. The calculation formula is: ; In the above formula, This represents the total number of original rate of change entries within the group. To eliminate the 10% extreme values, that is, the top 10% maximum values ​​and the bottom 10% minimum values, This represents the j-th original rate of change after sorting.

[0021] S2213, Based on the throttle reference temperature and the rate of change of temperature during the solar terms Construct a temperature prediction sub-model based on solar terms The expression is: ; In the above formula, For any time in the future With the current moment The time difference.

[0022] When the locomotive travels across regions, the model can automatically adjust the regional strategy weights based on latitude and longitude to adapt to the characteristics of high altitude, low temperature, and large diurnal temperature range. The construction steps of the region-based temperature prediction sub-model are as follows: S2221. Calculate the regional environmental reference temperature using the driver's cab temperature reference value under a standard scenario within a certain region. The standard scenario is normal vehicle speed and no extreme terrain. The calculation formula is: ; In the above formula, The baseline average temperature for route L. The continentality index, The average continentality index for all regions is typically 50, with a coefficient of 0.005 representing the correction percentage for every 1 unit deviation in continentality. The specific calculation steps for the regional environmental reference temperature are as follows: S22211. Calculate the average temperature of the driver's cab in a certain region based on a standard dataset. The calculation formula is: ; In the above formula, This represents the total number of valid data entries within region L. This is the i-th valid temperature data point within region L; S22212, Average value corrected by altitude To obtain the base average temperature For every 100m increase in altitude, the ambient temperature decreases by approximately 0.6℃, therefore the baseline average temperature needs to be adjusted. The calculation formula is: ; In the above formula, This represents the actual altitude of the current location. For reference altitude, the average altitude of the route is used, such as the Beijing-Shanghai line. =50m, coefficient 0.006 is the temperature correction per meter of altitude; S22213, Calculate the continentality index This reflects the degree to which a region is influenced by the continent, and the calculation formula is as follows: ; In the above formula, The annual temperature range for this region, which is the difference between the highest and lowest monthly average temperatures throughout the year, can be obtained from meteorological databases and can be queried through historical meteorological databases, such as those for Harbin. ≈42°, Haikou ≈12°; S22214, Based on the average temperature And Continental Index The regional environmental reference temperature was calculated. .

[0023] S2222. Calculate the rate of temperature change in a region by using the average rate of temperature change over time within that region. The calculation formula is: ; In the above formula, Let L be the average rate of change within the regional group of route L. This is a correction factor for outdoor temperature fluctuations. When the outdoor temperature changes drastically, the temperature change in the driver's cab also increases synchronously. The regional vehicle speed correction factor is used, as vehicle speed has a more significant impact on ventilation in high-altitude areas. The average rate of change within a regional group is the average rate of temperature change over time in a given region, reflecting how much the temperature rises or falls per hour. Its magnitude is strongly correlated with the continentality, altitude, and diurnal characteristics of the region. For example, the diurnal rate of change is higher inland than in coastal areas. The calculation steps for the average rate of change within a regional group are as follows: S22221. Filter the standard dataset to obtain continuous temperature sequences, and use the least squares method to calculate the original rate of change of a single sequence within the same region and time period group. The calculation formula is: ; In the above formula, This refers to the number of data points in the sequence; for example, a 2-hour sequence contains 24 data points. Relative time within the sequence, The relative temperature within the sequence; S22222. For all the original rates of change in the same region and time period group, remove the 10% extreme values ​​and then smooth them to obtain the average rate of change within the region group. The calculation formula is: ; In the above formula, This represents the total number of original rate of change entries within the group. To eliminate the 10% extreme values, we need the top 10% maximum values ​​and the bottom 10% minimum values. This represents the j-th original rate of change after sorting.

[0024] S2223, Based on regional environmental reference temperature and regional temperature change rate Constructing a region-based temperature prediction sub-model Its expression is: ; In the above formula, For any time in the future With the current moment The time difference.

[0025] S23, The temperature prediction sub-model based on solar terms and region-based temperature prediction sub-model By fusion, a temperature prediction model is obtained. Its expression is: ; In the above formula, The dynamic weight of the solar term at time t. Let t be the regional dynamic weight. The solar term dynamic weight is the weight that adjusts the contribution of the solar term to temperature prediction in real time according to the importance of the solar term. The regional dynamic weight is the weight that adjusts the contribution of the region to temperature prediction in real time according to the salience of regional characteristics and the degree of location change. Both are adapted to real-time scenarios to improve the accuracy of temperature prediction models. The calculation steps of the solar term dynamic weight and the regional dynamic weight are as follows: S231. Extracting solar terms from the calendar Solar Terms The quantification is 1-24, and the solar term intensity factor is calculated by combining the periodic characteristics of the solar terms. The formula for quantifying the basic impact of solar terms on temperature is as follows: ; In the above formula, The 24 solar terms The time difference between the current date and the S-th solar term. The cosine function of the intensity decay of the solar term. The basic importance weight of the Sth solar term reflects the degree of influence of this solar term, and its range is [0.6, 1.0]. S232. Derive the continentality index using latitude and longitude information, and then use the normalized continentality index... and altitude correction factor Comprehensive calculation of regional intensity factor Quantifying the uniqueness of regional climates, such as inland high-altitude areas, regional intensity factors. When the value is close to 1, regional strategy dominates; in coastal low-altitude areas, the regional intensity factor is... When the value approaches 0.2, the seasonal strategy becomes more important; the calculation formula is as follows: ; In the above formula, The normalized continentality index. Altitude correction factor, regional intensity factor The upper limit is set to 1 to avoid weight imbalance caused by extreme altitudes. S233, Regarding the intensity factor of solar terms and regional intensity factor Normalization is performed to generate initial weights, and the expressions for both are as follows: ; In the above formula, The interaction coefficient, ranging from 0.1 to 0.3, is determined through training with historical data and is used to correct weight biases in scenarios where solar terms are strongly correlated with geographical regions, such as coastal areas in summer. =0.8, =0.2, =0.2, =3, at this point the calculation yields... ≈0.8, ≈0.2, at this time the solar term takes precedence, in inland high-altitude areas in spring: =0.4, =0.9, =0.2, =2, at this point the calculation yields... ≈0.3, ≈0.7, at this point the region is the primary factor; S234. Combining real-time prediction errors and regional change rates, optimize the initial weights to obtain dynamic weights for each solar term. and regional dynamic weights The initial weights need to be dynamically adjusted based on real-time locomotive operating data to ensure that the weights are optimized in real time as the environment changes. The expressions for the two are as follows: ; In the above formula, The function restricts the weights to a range of 0.1-0.9 to prevent a single strategy from being completely ignored. To adjust the rate, a value of 0.05-0.2 is used to avoid sudden changes in weights. To find the minimum value, to prevent the denominator from being 0. Temperature prediction sub-model based on solar terms The prediction error Location-based temperature prediction submodel The prediction error.

[0026] S24. Temperature prediction model using standard dataset To perform training and obtain the trained model parameters and The trained temperature prediction model is obtained. The objective function for model training is: ; In the above formula, Let T be the baseline value for the weight of the solar term. Let T be the baseline value for the regional weight. The actual temperature values ​​at time t during the S solar term and along route L. To optimize the objective function that compares the differences between the dynamic weights of solar terms and regional dynamic weights and the benchmark weights, The objective function is the difference between the model's predicted temperature and the baseline predicted temperature. By constraining the two objective functions, the parameters of the temperature prediction model used for subsequent calculations can be obtained.

[0027] S3. Input the actual data samples into the trained temperature prediction model, and output the predicted temperature change trend of the driver's cab within the preset time period. The specific process of the above steps is as follows: First, obtain the average actual temperature of the driver's cab within this solar term using actual data samples, and then adjust the model according to the climate correction coefficient. The actual line-level reference temperature correction factor is calculated, and the actual throttle reference temperature is calculated using the actual line-level reference temperature correction factor. Then, based on the throttle temperature change rate... The actual predicted temperature based on the solar terms is calculated; then, the average actual temperature in the driver's cab within the region is obtained through actual data samples, and the temperature is calculated according to the continentality index. The actual regional environmental reference temperature is calculated, and the actual regional reference temperature is obtained from the actual regional environmental reference temperature, and then based on the regional temperature change rate. The actual predicted temperature based on the region is calculated, and the actual predicted temperature based on the solar term and the actual predicted temperature based on the region are substituted into the temperature prediction model. This will provide a prediction of the temperature change trend over a period of time.

[0028] S4. Compare the predicted temperature change trend with the preset target temperature range for the driver's cab to determine whether the future temperature deviates from the target temperature range. The predicted temperature change trend of the driver's cab output in step S3 within the preset time period, for example, if the temperature will rise from 25°C to 30°C in the next 30 minutes, exceeding the target range of 22-26°C, will be directly transmitted to the controller. Based on the comparison between the predicted result and the preset target temperature range, the controller decides whether to perform the adjustment action and how to adjust it. If so, proceed to step S5. If the prediction result shows that the future temperature will deviate from the target range, the controller will generate targeted adjustment instructions based on the predicted deviation range, deviation time and other details, such as the temperature will reach 31°C in 10 minutes, exceeding the upper limit by 5°C, such as starting the air conditioner to cool down and adjusting the power to 70%. If not, the controller will not generate adjustment commands, and the prediction results will show that the future temperature will always be within the target range; S5. The controller generates the corresponding temperature adjustment command and sends it to the temperature actuator. The temperature actuator receives the temperature adjustment command and performs the corresponding temperature adjustment action to maintain the temperature in the driver's cab within the target temperature range. The generated command must be encoded according to a preset protocol (such as the CAN bus communication protocol) to ensure that the temperature actuator can parse it. The controller sends the encoded adjustment command to the temperature actuator through a wired / wireless communication module. After the actuator parses the valid command, it drives the hardware device (such as the air conditioning compressor) to perform the corresponding action.

[0029] Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0030] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. Since the above embodiments are substantially similar to the method embodiments, their descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0031] The above embodiments provide a detailed description of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for adaptive temperature control in a railway locomotive driver's cab, characterized in that, The adjustment method includes the following steps: S1. Divide the target area into several target regions, and sequentially obtain several historical data samples of each target region under each target solar term. The historical data samples include historical indoor and outdoor temperatures of the driver's cab, locomotive operating speed, solar term labels, and latitude and longitude information. Obtain actual data samples of the current target region under the current target solar term through the information acquisition module. The actual data samples include actual indoor and outdoor temperatures of the driver's cab, actual locomotive operating speed, solar term labels, and latitude and longitude information. S2. Construct a temperature prediction model that integrates the temperature regulation strategy based on the solar terms and the temperature regulation strategy based on the region, and optimize the model parameters through iterative training using historical data samples to obtain the trained temperature prediction model. S3. Input the actual data sample into the trained temperature prediction model and output the predicted temperature change trend of the driver's cab within a preset time period. S4. Compare the predicted temperature change trend with the preset target temperature range for the driver's cab to determine whether the future temperature deviates from the target temperature range. If so, proceed to step S5; If not, the controller will not generate an adjustment command; S5. The controller generates the corresponding temperature adjustment command and sends it to the temperature actuator. The temperature actuator receives the temperature adjustment command and executes the corresponding temperature adjustment action to maintain the temperature in the driver's cab within the target temperature range.

2. The adaptive temperature control method for railway locomotive driver's cab according to claim 1, characterized in that, In step S2, the steps for constructing the temperature prediction model are as follows: S21. Clean the historical data samples to generate a standard dataset as training samples. S22. Construct a temperature prediction sub-model based on solar terms using standard datasets. and region-based temperature prediction sub-model ; S23, The temperature prediction sub-model based on solar terms and region-based temperature prediction sub-model By fusion, a temperature prediction model is obtained. Its expression is: ； In the above formula, The dynamic weight of the solar term at time t. Let be the regional dynamic weight at time t; S24. Temperature prediction model using standard dataset To perform training and obtain the trained model parameters and The trained temperature prediction model is obtained. The objective function for model training is: ； In the above formula, Let T be the baseline value for the weight of the solar term. Let T be the baseline value for the regional weight. The actual temperature values ​​at time t during the S solar term and along route L. To optimize the objective function that compares the differences between the dynamic weights of solar terms and regional dynamic weights and the benchmark weights, The objective function is the difference between the model-predicted temperature and the baseline-predicted temperature.

3. The adaptive temperature control method for railway locomotive driver's cab according to claim 2, characterized in that, In step S22, the construction steps of the temperature prediction sub-model based on the solar terms are as follows: S2211. Calculate the reference temperature for a given solar term using the reference value of the driver's cab temperature under a standard scenario within that solar term. The calculation formula is: ； In the above formula, This is the line-level reference temperature correction factor for route L during throttle S. For vehicle model correction; S2212. Calculate the temperature change rate of a solar term based on the average rate of temperature change over time within that solar term. The calculation formula is: ； In the above formula, This represents the average rate of change within the same solar term and time period group. This is the solar radiation correction factor. This is the vehicle speed correction factor; S2213, via throttle reference temperature and the rate of change of temperature during the solar terms Construct a temperature prediction sub-model based on solar terms The expression is: ； In the above formula, For any time in the future With the current moment The time difference.

4. The adaptive temperature control method for railway locomotive driver's cab according to claim 3, characterized in that, In step S2211, the calculation steps for the line-level reference temperature correction coefficient are as follows: S22111. Calculate the average temperature in the driver's cab during a certain throttle period based on a standard dataset. The calculation formula is: ； In the above formula, This represents the total number of valid data entries within solar term S. This is the i-th valid temperature data point within solar term S; S22112, Calculation of Line Climate Correction Coefficient Based on Line Outdoor Temperature The calculation formula is: ； In the above formula, Let L be the average outdoor temperature of line L during the solar term S. The average outdoor temperature for all lines during solstice S; S22113, The average temperature of the driver's cab during a certain solar term. With line climate correction factor The line-level reference temperature correction factor was obtained by calculation. The calculation formula is: 。 5. The adaptive temperature control method for railway locomotive driver's cab according to claim 3, characterized in that, In step S2212, the calculation steps for the intra-group average rate of change of the same solar term and time period are as follows: S22121. Filter the standard dataset to obtain continuous temperature sequences, and use the least squares method to calculate the original rate of change of a single sequence within the same solar term and time period group. The calculation formula is: ； In the above formula, The number of data points in the sequence. Relative time within the sequence, The relative temperature within the sequence; S22122. For all the original rates of change in the same solar term and time period group, after removing the 10% extreme values, the average rate of change within the group is obtained through smoothing. The calculation formula is: ； In the above formula, This represents the total number of original rate of change entries within the group. To eliminate the 10% extreme values, This represents the j-th original rate of change after sorting.

6. The adaptive temperature control method for a railway locomotive driver's cab according to claim 2, characterized in that, In step S22, the construction steps of the region-based temperature prediction sub-model are as follows: S2221. Calculate the regional environmental reference temperature using the driver's cab temperature reference value under a standard scenario within a certain region. The calculation formula is: ； In the above formula, The baseline average temperature for route L. The continentality index, The average continentality index for all regions, with a coefficient of 0.005 representing the correction ratio for every 1 unit deviation in continentality; S2222. Calculate the rate of temperature change in a region by using the average rate of temperature change over time within that region. The calculation formula is: ； In the above formula, Let L be the average rate of change within the regional group of route L. This is a correction factor for outdoor temperature fluctuations. This is a regional vehicle speed correction factor; S2223, Based on regional environmental reference temperature and regional temperature change rate Constructing a region-based temperature prediction sub-model Its expression is: ； In the above formula, For any time in the future With the current moment The time difference.

7. The adaptive temperature control method for a railway locomotive driver's cab according to claim 6, characterized in that, In step S2221, the specific implementation steps are as follows: S22211. Calculate the average temperature of the driver's cab in a certain region based on a standard dataset. The calculation formula is: ； In the above formula, This represents the total number of valid data entries within region L. This is the i-th valid temperature data point within region L; S22212, Average value corrected by altitude To obtain the baseline average temperature The calculation formula is: ； In the above formula, This represents the actual altitude of the current location. For reference altitude, a coefficient of 0.006 represents the temperature correction per meter of altitude. S22213, Calculate the continentality index The calculation formula is: ； In the above formula, This refers to the annual temperature range in this region; S22214, Based on the average temperature And Continental Index The regional environmental reference temperature was calculated. .

8. The adaptive temperature control method for a railway locomotive driver's cab according to claim 6, characterized in that, In step S2222, the calculation steps for the average rate of change within the regional group are as follows: S22221. Filter the standard dataset to obtain continuous temperature sequences, and use the least squares method to calculate the original rate of change of a single sequence within the same region and time period group. The calculation formula is: ； In the above formula, The number of data points in the sequence. Relative time within the sequence, The relative temperature within the sequence; S22222. For all the original rates of change in the same region and time period group, remove the 10% extreme values ​​and then smooth them to obtain the average rate of change within the region group. The calculation formula is: ； In the above formula, This represents the total number of original rate of change entries within the group. To eliminate the 10% extreme values, This represents the j-th original rate of change after sorting.

9. The adaptive temperature control method for a railway locomotive driver's cab according to claim 2, characterized in that, In step S23, the calculation steps for the solar term dynamic weight and the regional dynamic weight are as follows: S231. Extracting solar terms from the calendar Calculate the solar term intensity factor by combining the periodic characteristics of the solar terms. The calculation formula is: ； In the above formula, The 24 solar terms The time difference between the current date and the S-th solar term. The cosine function of the intensity decay of the solar term. The basic importance weight of the Sth solar term; S232. Derive the continentality index using latitude and longitude information, and then use the normalized continentality index... and altitude correction factor Comprehensive calculation of regional intensity factor The calculation formula is: ； In the above formula, The normalized continentality index. Altitude correction factor; S233, Regarding the intensity factor of solar terms and regional intensity factor Normalization is performed to generate initial weights, and the expressions for both are as follows: ； In the above formula, The interaction coefficient; S234. Combining real-time prediction errors and regional change rates, optimize the initial weights to obtain dynamic weights for each solar term. and regional dynamic weights The expressions for the two are as follows: ； In the above formula, The function's weights are restricted to a range of 0.1-0.

9. To adjust the rate, It is the minimum value. Temperature prediction sub-model based on solar terms The prediction error Location-based temperature prediction submodel The prediction error.

Images