Intelligent Temperature Control Method for Cigar Cabinets Based on Multi-Sensor Fusion

Abstract

This invention discloses an intelligent temperature control method for cigar cabinets based on multi-sensor fusion, belonging to the field of cigar care equipment control technology. The method includes deploying a sensor array to periodically collect environmental data within the cabinet; constructing a multi-dimensional distributed field model based on the data and calculating thermodynamic coupling characteristic values; comparing the characteristic values ​​with preset thresholds to generate a preliminary control strategy; combining historical compressor load data analysis to generate a final control command sequence containing dynamic timing; and driving the actuator to perform actions. This method accurately characterizes the three-dimensional spatial environment state through a field model, realizing a shift from point-based control to overall uniform control; and optimizes the compressor start-stop rhythm through timing-based commands, effectively reducing equipment wear and energy consumption while ensuring precise temperature control, extending equipment life, and maintaining a stable care environment.

Description

Technical Field

[0001] This invention belongs to the field of cigar maintenance equipment control technology, specifically a smart temperature control method for cigar cabinets based on multi-sensor fusion. Background Technology

[0002] Traditional cigar cabinet temperature and humidity control relies primarily on monitoring by sensors at single or a few fixed points. This point-based measurement method treats the collected local parameters as the average state of the entire storage cavity and performs simple on / off or proportional-integral-derivative feedback control accordingly. However, the interior space of a cigar cabinet typically exhibits microenvironmental differences due to factors such as uneven airflow, product placement, and the location of the cooling source. A single measurement point cannot accurately reflect the distribution gradient and non-uniformity of environmental parameters throughout the three-dimensional space. Current technology's understanding of the cigar care environment remains at the level of "points" or "average surfaces," lacking precise modeling of the overall physical field state within the cavity.

[0003] Existing control strategies typically only consider the instantaneous deviation between current environmental parameters and set target values ​​when generating adjustment commands, directly driving actuators such as compressors to achieve rapid correction. This instantaneous response mode ignores the historical operating state and thermal inertia of the actuators, especially the compressor. Frequent start-stop commands based on instantaneous deviations can easily lead to short-cycle operation of the compressor, generating large impact loads, which not only increases energy consumption but also accelerates mechanical wear, affecting its long-term operational reliability and lifespan. At the same time, drastic adjustment actions may also disturb the stability of airflow within the cabinet, which is detrimental to the smooth and gradual environmental changes required for cigar curing. Therefore, a refined control method that can comprehensively consider the spatial environmental field state and the operating history of the actuators is needed to resolve the contradiction between microenvironmental imbalances and equipment wear. Summary of the Invention

[0004] This invention aims to solve at least one of the technical problems existing in the prior art;

[0005] Therefore, this invention proposes a smart temperature control method for cigar cabinets based on multi-sensor fusion, comprising:

[0006] Temperature sensor array, humidity sensor array and airflow pressure sensor array are deployed inside the cigar cabinet to periodically collect raw environmental parameter data of the cigar cabinet's internal cavity;

[0007] Based on the original environmental parameter data, a multidimensional distribution field model of the internal cavity of the cigar cabinet is constructed, and the thermodynamic coupling characteristic value characterizing the current environmental state is calculated through the multidimensional distribution field model.

[0008] The thermodynamic coupling eigenvalues ​​are compared with the eigenvalue thresholds of the preset optimal cigar maintenance model, and a preliminary temperature regulation strategy is generated based on the comparison results.

[0009] Based on the analysis of the preliminary temperature regulation strategy and the historical workload data of the cigar cabinet compressor, a final temperature regulation command sequence containing dynamic execution timing is generated;

[0010] The final temperature adjustment command sequence is sent to the temperature actuator of the cigar cabinet, driving the temperature actuator to perform the corresponding temperature adjustment action.

[0011] Furthermore, the deployment of temperature sensor arrays, humidity sensor arrays, and airflow pressure sensor arrays inside the cigar cabinet to periodically collect raw environmental parameter data of the cigar cabinet's internal cavity includes:

[0012] The temperature sensor array, humidity sensor array, and airflow pressure sensor array are respectively arranged in the upper, middle, and lower three layers of the cigar cabinet and in key locations such as near the door.

[0013] According to the preset sampling frequency, the temperature data point set output by the temperature sensor array, the humidity data point set output by the humidity sensor array, and the pressure data point set output by the airflow pressure sensor array are read and recorded synchronously.

[0014] The temperature data point set, humidity data point set, and pressure data point set are timestamped and combined to form the original environmental parameter data.

[0015] Furthermore, the construction of a multidimensional distribution field model of the cigar cabinet's internal cavity based on the original environmental parameter data, and the calculation of thermodynamic coupling characteristic values ​​representing the current environmental state using the multidimensional distribution field model, includes:

[0016] The temperature, humidity, and pressure data points, after being aligned with the timestamps, are interpolated and fitted onto the three-dimensional spatial coordinates of the cigar cabinet's internal cavity to generate continuous temperature, humidity, and pressure distribution fields, respectively.

[0017] Analyze the gradient vector relationship between the continuous temperature distribution field and the continuous humidity distribution field, and calculate the inter-field cooperative change index;

[0018] Analyze the influence of the continuous pressure distribution field on the energy transfer of the temperature and humidity fields, and calculate the heat and mass transfer resistance coefficient;

[0019] By integrating the inter-field synergistic change index and the heat and mass transfer resistance coefficient, the thermodynamic coupling characteristic value is output through the characteristic value calculation engine.

[0020] Furthermore, the step of comparing the thermodynamic coupling feature value with the feature value threshold of the preset optimal cigar maintenance model, and generating a preliminary temperature regulation strategy based on the comparison result, includes:

[0021] The optimal cigar maintenance model includes a range of feature value thresholds corresponding to an ideal maintenance environment.

[0022] Determine whether the thermodynamic coupling characteristic value falls within the characteristic value threshold range;

[0023] If the thermodynamic coupling eigenvalue is lower than the lower limit of the eigenvalue threshold range, a temperature enhancement strategy aimed at improving overall thermodynamic activity is generated.

[0024] If the thermodynamic coupling eigenvalue is higher than the upper limit of the eigenvalue threshold range, a cooling suppression strategy aimed at reducing overall thermodynamic activity is generated.

[0025] If the thermodynamic coupling characteristic value falls within the characteristic value threshold range, a isothermal maintenance strategy to maintain the current state is generated.

[0026] The heating enhancement strategy, cooling suppression strategy, or isothermal maintenance strategy are collectively referred to as the preliminary temperature regulation strategy.

[0027] Furthermore, the generation of a final temperature regulation command sequence containing dynamic execution timing, based on the analysis of the preliminary temperature regulation strategy and the historical workload data of the cigar cabinet compressor, includes:

[0028] Extract historical workload data of the compressor within a preset time period from the historical log of the cigar cabinet control unit, including compressor start frequency, single run duration, and power change curve;

[0029] Analyze the periodic patterns and load intensity trends of the compressor's historical operating load data;

[0030] Based on the target intensity of the preliminary temperature regulation strategy, assess the potential load impact on the compressor that direct execution of the preliminary temperature regulation strategy may cause;

[0031] Based on the load impact assessment results, the objectives of the preliminary temperature regulation strategy are decomposed into multiple gradient sub-objectives, and execution timing and intensity parameters are configured for each gradient sub-objective to form the final temperature regulation command sequence.

[0032] Furthermore, the assessment of the potential load impact on the compressor that might result from directly implementing the initial temperature regulation strategy includes:

[0033] Based on the target temperature change of the preliminary temperature regulation strategy, calculate the required incremental compressor energy consumption.

[0034] The estimated increase in compressor energy consumption is compared with the recent average load represented in the compressor's historical operating load data to calculate the load impact coefficient.

[0035] The load impact level is determined based on the predefined range in which the load impact coefficient falls.

[0036] Furthermore, the step of decomposing the objective of the initial temperature regulation strategy into multiple gradient sub-objectives based on the load shock assessment results includes:

[0037] When the load shock assessment result is a high shock level, the target temperature change of the preliminary temperature regulation strategy is decomposed into four or more incremental gradient sub-objectives.

[0038] When the load shock assessment result is a medium shock level, the target temperature change of the preliminary temperature regulation strategy is decomposed into two to three step-gradient sub-targets.

[0039] When the load shock assessment result is a low shock level, the target temperature change of the preliminary temperature regulation strategy is taken as a gradient sub-objective.

[0040] Furthermore, after sending the final temperature adjustment command sequence to the temperature actuator of the cigar cabinet and driving the temperature actuator to perform the corresponding temperature adjustment action, the method further includes:

[0041] During the execution of the temperature regulation action, the original environmental parameter data are continuously collected and updated;

[0042] Based on the updated original environmental parameter data, new thermodynamic coupling characteristic values ​​are calculated in real time.

[0043] Monitor the dynamic response rate at which the new thermodynamic coupling eigenvalue approaches the eigenvalue threshold of the preset optimal cigar maintenance model;

[0044] Based on the dynamic response rate, the timing and intensity parameters of subsequent unexecuted gradient sub-objectives are fine-tuned.

[0045] Furthermore, the calculation of the heat and mass transfer resistance coefficient includes:

[0046] Based on the gradient of the continuous pressure distribution field, the main airflow path and static pressure zone of the cigar cabinet's internal cavity are identified;

[0047] Based on the principles of fluid mechanics, calculate the theoretical mass transfer efficiency along the main path of the airflow.

[0048] The ratio of the theoretical mass transfer efficiency to the actual mass transfer efficiency derived from changes in the temperature and humidity fields is defined as the heat and mass transfer resistance coefficient.

[0049] Furthermore, the step of fine-tuning the execution timing and intensity parameters of subsequent unexecuted gradient sub-objectives based on the dynamic response rate includes:

[0050] If the dynamic response rate is faster than the preset expected response rate, the execution timing of subsequent gradient sub-objectives will be appropriately delayed or their intensity parameters will be reduced.

[0051] If the dynamic response rate is slower than the preset expected response rate, the execution timing of subsequent gradient sub-objectives should be advanced or their intensity parameters should be increased.

[0052] If the dynamic response rate matches the preset expected response rate, then the preset execution timing and intensity parameters in the final temperature adjustment command sequence remain unchanged.

[0053] Compared with the prior art, the beneficial effects of the present invention are:

[0054] An array of temperature, humidity, and airflow pressure sensors is deployed, and a multidimensional distribution field model is constructed based on the spatiotemporal data collected, calculating thermodynamic coupling characteristic values. This method elevates the control basis from discrete point measurements to a continuous field-state quantitative characterization of the interaction between the temperature, humidity, and airflow pressure fields within the cigar cabinet. This enables the system to sense and respond to uneven distribution of the environment in three-dimensional space and local hot / cold spots, identifying potential instability states that cannot be detected by simply relying on average temperature or humidity. By comparing this comprehensive characteristic value with the target threshold, the generated preliminary adjustment strategy can guide the system to perform environmental adjustments aimed at improving the uniformity of the entire spatial field, rather than just correcting parameters at a single point, from the perspective of overall thermodynamic balance. This provides a precise control benchmark for achieving comprehensive and balanced cigar maintenance.

[0055] Based on the initial adjustment strategy, historical compressor workload data is analyzed to generate a final temperature control command sequence containing dynamic execution timing. This process couples the long-term operational health of the actuator with the immediate environmental control requirements. By analyzing historical loads, the system can determine the compressor's working inertia and fatigue level, thereby optimizing the initial strategy in the time dimension. The generated command sequence plans the specific timing, intensity, and sequence of actions such as compressor start-up and shutdown, and fan speed adjustment. Its core objective is to make the environmental control process smooth and gradual. This avoids frequent compressor start-ups and shutdowns and load shocks caused by pursuing rapid correction, reducing equipment operating stress and energy consumption, and extending the service life of core components. At the same time, the gentle adjustment actions also help maintain the stability of airflow and temperature and humidity changes within the cabinet, preventing environmental disturbances caused by drastic adjustments, which is more in line with the cigar's need for a slow and stable aging environment. Attached Figure Description

[0056] Figure 1 This is a flowchart illustrating the steps of the intelligent temperature control method for a cigar cabinet based on multi-sensor fusion as described in this invention.

[0057] Figure 2 A flowchart for collecting raw environmental parameter data;

[0058] Figure 3 A flowchart for calculating thermodynamic coupling eigenvalues;

[0059] Figure 4 Comparison of key parameters of adjustment strategies under different load impact levels;

[0060] Figure 5 This is a dynamic thermal map showing the temperature distribution across multiple layers and zones of a cigar cabinet. Detailed Implementation

[0061] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. 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.

[0062] See Figure 1 A temperature sensor array, a humidity sensor array, and an airflow pressure sensor array are deployed inside the cigar cabinet to periodically collect raw environmental parameter data of the cigar cabinet's internal cavity. Based on the raw environmental parameter data, a multidimensional distribution field model of the cigar cabinet's internal cavity is constructed. The thermodynamic coupling characteristic value characterizing the current environmental state is calculated through the multidimensional distribution field model. The thermodynamic coupling characteristic value is compared with the characteristic value threshold of the preset optimal cigar maintenance state model. Based on the comparison results, a preliminary temperature regulation strategy is generated. Based on the analysis of the preliminary temperature regulation strategy and the historical workload data of the cigar cabinet compressor, a final temperature regulation command sequence containing dynamic execution timing is generated. The final temperature regulation command sequence is sent to the temperature actuator of the cigar cabinet to drive the temperature actuator to perform the corresponding temperature regulation action.

[0063] See Figure 2In one embodiment of the present invention, the internal space of the cigar cabinet is divided into three layers: upper, middle, and lower, as well as a region near the door. Temperature sensor arrays, humidity sensor arrays, and airflow pressure sensor arrays are respectively arranged at key locations on each layer. The temperature sensor array includes multiple discretely distributed temperature measuring units, the humidity sensor array includes multiple discretely distributed humidity measuring units, and the airflow pressure sensor array includes multiple discretely distributed pressure sensing units. For example, the upper layer is equipped with three temperature sensing units, two humidity sensing units, and one pressure sensing unit; the middle layer is equipped with four temperature sensing units, three humidity sensing units, and two pressure sensing units; the lower layer is equipped with three temperature sensing units, two humidity sensing units, and one pressure sensing unit; and the region near the door is equipped with two temperature sensing units, one humidity sensing unit, and one pressure sensing unit. These sensing units are installed on the inner wall of the cigar cabinet with fixed spatial coordinates, forming a three-dimensional monitoring network covering the internal cavity.

[0064] In some embodiments, temperature data points output by the temperature sensor array, humidity data points output by the humidity sensor array, and pressure data points output by the airflow pressure sensor array are synchronously read and recorded according to a preset sampling frequency. The preset sampling frequency is set to once per minute. Each temperature measuring unit of the temperature sensor array outputs a temperature value at the sampling time, and the temperature values ​​output by all temperature measuring units constitute the temperature data point set. Each humidity measuring unit of the humidity sensor array outputs a humidity value at the sampling time, and the humidity values ​​output by all humidity measuring units constitute the humidity data point set. Each pressure sensing unit of the airflow pressure sensor array outputs a pressure value at the sampling time, and the pressure values ​​output by all pressure sensing units constitute the pressure data point set. The synchronous reading process is coordinated by the central control module embedded in the cigar cabinet. The central control module sends a synchronous trigger signal to all sensing units to ensure that the temperature data point set, humidity data point set, and pressure data point set correspond to the same sampling time. The recorded data point set is accompanied by timestamp information, which includes year, month, day, hour, minute, and second.

[0065] In practice, the temperature, humidity, and pressure data points are timestamped and combined to form the original environmental parameter data. Timestamp alignment is based on the minimum time difference matching principle. For each sampling moment, the central control module checks the timestamps of the temperature, humidity, and pressure data points. If there is a slight deviation in the timestamps, an interpolation method is used for alignment. The interpolation method uses a linear interpolation formula to calculate the aligned data values. The linear interpolation formula is expressed as:

[0066]

[0067] in: These are the aligned data values. It is the target alignment timestamp. and It is the actual collection timestamp and The two closest timestamps and It corresponds to and The data values, the aligned temperature data point set, humidity data point set, and pressure data point set are arranged in chronological order and combined to form the original environmental parameter data. The original environmental parameter data is a multidimensional dataset containing a timestamp sequence, a spatial coordinate sequence, and the corresponding temperature value sequence, humidity value sequence, and pressure value sequence.

[0068] It is understandable that the timestamp alignment process ensures the temporal consistency of the original environmental parameter data. Data comparison shows that unaligned temperature and humidity data points may have a maximum time deviation of 500 milliseconds at the same sampling time. After alignment, the deviation is eliminated, and the original environmental parameter data is used for subsequent modeling and analysis. In some embodiments, the preset sampling frequency can be adjusted according to the cigar cabinet's operating mode. For example, in rapid adjustment mode, the sampling frequency is increased to once every 30 seconds, and in stable maintenance mode, the sampling frequency is reduced to once every 5 minutes. The adjustment of the sampling frequency is achieved through the configuration parameters of the central control module. The central control module dynamically switches the sampling frequency, while the synchronous reading mechanism remains unchanged. Optionally, the number of sensing units in the temperature sensor array, humidity sensor array, and airflow pressure sensor array can be increased or decreased. Increasing the number of sensing units increases the spatial monitoring density, while decreasing the number of sensing units reduces system complexity. Changes in the number of sensing units do not alter the basic process of synchronous reading and timestamp alignment. It is understandable that the combination of raw environmental parameter data is in the form of a structured data table. Each row of the structured data table represents a sampling time, and each column represents the data of a sensing unit. The structured data table is stored in the local memory of the cigar cabinet for subsequent processing modules to call.

[0069] See Figure 3In one embodiment of the present invention, the construction of the multidimensional distribution field model and the calculation of thermodynamic coupling eigenvalues ​​are performed. The temperature data point set, humidity data point set, and pressure data point set after time stamp alignment are interpolated and fitted on the three-dimensional spatial coordinates of the cigar cabinet's internal cavity. The three-dimensional spatial coordinates of the cigar cabinet's internal cavity are established in a rectangular coordinate system with one of the cabinet's interior corners as the origin. Each temperature value in the temperature data point set is associated with a three-dimensional spatial coordinate. Based on the spatial coordinates and values ​​of all temperature data points, a continuous temperature distribution field is generated using a cubic spline interpolation algorithm. The continuous humidity distribution field and the continuous pressure distribution field are generated using the same cubic spline interpolation algorithm based on their respective data point sets. The interpolated continuous temperature distribution field, continuous humidity distribution field, and continuous pressure distribution field are continuous scalar functions defined on the three-dimensional space of the cigar cabinet's internal cavity. Data comparison shows that the original temperature data point set contains twelve discrete data points, and the continuous temperature distribution field generated by interpolation and fitting can provide theoretical temperature values ​​for any spatial location.

[0070] In some embodiments, the gradient vector relationship between the continuous temperature distribution field and the continuous humidity distribution field is analyzed, and the inter-field cooperative change index is calculated. The gradient vector of the continuous temperature distribution field represents the direction and rate of the fastest temperature change at each point in space, and the gradient vector of the continuous humidity distribution field represents the direction and rate of the fastest humidity change at each point in space. For a series of uniform sampling points in space, the dot product of the temperature gradient vector and the humidity gradient vector at each sampling point is calculated. The average value of the dot product results for all sampling points is taken to obtain the inter-field cooperative change index. The formula for calculating the inter-field cooperative change index is as follows:

[0071]

[0072] in: This represents the inter-field coordinated change index. This represents the total number of spatially uniform sampling points. Indicates the first Temperature gradient vector at each sampling point Indicates the first The humidity gradient vector at each sampling point, denoted by . The dot product operation of vectors is represented by the field co-variance index, which is a scalar whose magnitude reflects the degree of consistency between the temperature field and the humidity field in spatial variation.

[0073] In practical implementation, the influence of the continuous pressure distribution field on the energy transfer of the temperature and humidity fields is analyzed, and the heat and mass transfer resistance coefficient is calculated. Based on the gradient of the continuous pressure distribution field, the main airflow path and static pressure zone inside the cigar cabinet are identified. The direction of the pressure gradient vector points to the direction of the fastest pressure decrease. Continuous areas with large pressure gradients are identified as the main airflow path, and areas with pressure gradients close to zero are identified as static pressure zones. The theoretical mass transfer efficiency on the main airflow path is calculated according to fluid mechanics principles. The theoretical mass transfer efficiency is calculated based on the laminar flow model and pressure difference. At the same time, the actual mass transfer efficiency is inferred based on the changes in the continuous temperature and humidity distribution fields within a preset time period. The theoretical mass transfer efficiency is compared with the actual mass transfer efficiency inferred based on the changes in the temperature and humidity fields, and the ratio is defined as the heat and mass transfer resistance coefficient. Data comparison shows that in one system operation, the calculated theoretical mass transfer efficiency is 0.85, and the inferred actual mass transfer efficiency is 0.68. The ratio of 0.8 is the heat and mass transfer resistance coefficient.

[0074] In one embodiment of the present invention, the comparison of thermodynamic coupling characteristic values ​​and the generation of preliminary strategies are carried out. The optimal state model for cigar maintenance includes a characteristic value threshold range corresponding to the ideal maintenance environment. The characteristic value threshold range is pre-stored in the configuration file of the cigar cabinet control system in the form of a numerical range. For example, the lower limit of the characteristic value threshold range is 0.8 and the upper limit of the characteristic value threshold range is 1.2. The judgment step compares the calculated real-time thermodynamic coupling characteristic value with the characteristic value threshold range. The data comparison shows that the thermodynamic coupling characteristic value calculated in one system operation cycle is 0.75, which is lower than the preset lower limit of the characteristic value threshold range of 0.8.

[0075] In some embodiments, the logic for determining whether a thermodynamic coupling eigenvalue falls within an eigenvalue threshold range is implemented through a mathematical comparison expression, which is: or or ,in This represents the thermodynamic coupling eigenvalue obtained from the current calculation. This represents the lower limit of the eigenvalue threshold range. This represents the upper limit of the characteristic value threshold range. The control system sequentially calculates the truth value of the logical expressions. If the expression... If true, then the thermodynamic coupling eigenvalue is determined to be below the lower limit of the eigenvalue threshold range. If the expression... If true, then the thermodynamic coupling eigenvalue is determined to be higher than the upper limit of the eigenvalue threshold range. If the expression... If true, then the thermodynamic coupling eigenvalue is determined to fall within the eigenvalue threshold range.

[0076] In specific implementation, if the thermodynamic coupling characteristic value is lower than the lower limit of the characteristic value threshold range, a temperature enhancement strategy aimed at improving overall thermodynamic activity is generated. The temperature enhancement strategy is a set of instructions including a target temperature setpoint, a compressor start command, and an expected adjustment duration. For example, the target temperature setpoint is 2 degrees Celsius higher than the current average temperature inside the cabinet. If the thermodynamic coupling characteristic value is higher than the upper limit of the characteristic value threshold range, a cooling suppression strategy aimed at reducing overall thermodynamic activity is generated. The cooling suppression strategy includes a lower target temperature setpoint and a fan auxiliary ventilation command. If the thermodynamic coupling characteristic value falls within the characteristic value threshold range, a constant temperature holding strategy is generated to maintain the current state. The constant temperature holding strategy's instruction is to maintain the current operating state of the compressor and all auxiliary actuators unchanged.

[0077] It is understood that the heating enhancement strategy, cooling suppression strategy, or isothermal maintenance strategy are collectively referred to as the preliminary temperature regulation strategy. This preliminary temperature regulation strategy exists in the system memory as a structured data object, containing a strategy type identifier and specific control parameter key-value pairs. In some embodiments, the lower and upper limits of the characteristic value threshold range can be calibrated and set via a user interface. The user adjusts the values ​​of the lower limit L and upper limit U through the touchscreen input interface of the cigar cabinet, based on the specific type of cigars stored. The adjusted values ​​are then updated to the configuration file. Optionally, the judgment logic can be extended to include a hysteresis comparison with a buffer, for example, when a thermodynamically coupled characteristic value enters the interval [L, U] from below, the comparison continues until the characteristic value reaches... (in Only when the value is a small positive number is it considered to fall within the range, in order to avoid frequent policy switching near the threshold boundary. It can be understood that the process of generating a preliminary temperature regulation policy is a deterministic logical mapping process, and the generation of the policy is entirely determined by the comparison relationship between the thermodynamic coupling eigenvalue and the preset eigenvalue threshold range.

[0078] In one embodiment of the present invention, historical compressor workload data for a preset time period is extracted from the historical log of the cigar cabinet control unit. The preset time period is the most recent 24 hours. The historical compressor workload data includes compressor start frequency, single run duration, and power change curve. The compressor start frequency is the number of times the compressor enters the running state from a stopped state per unit time. The single run duration is the duration of each continuous operation of the compressor. The power change curve records the sequence data of the compressor's power change over time during each operation. The extracted data is stored in a time series format. The periodicity and load intensity trend of the historical compressor workload data are analyzed. The periodicity is identified by analyzing the distribution of compressor start events on the time axis, for example, by Fourier transform or by finding patterns in start intervals. The load intensity trend is obtained by calculating the direction of change of the total energy consumption or average power of the compressor operating per unit time, for example, analyzing whether the average hourly energy consumption of the most recent six hours has increased, decreased, or remained the same compared to the previous six hours. The target strength of the initial temperature regulation strategy is assessed in conjunction with the potential load impact on the compressor caused by directly implementing the initial temperature regulation strategy. The target strength of the initial temperature regulation strategy is determined by both the strategy type and the target temperature change. For example, the target of an initial temperature enhancement strategy is to raise the average temperature inside the cabinet by 2.5 degrees Celsius.

[0079] In practical implementation, assessing the potential load impact on the compressor caused by directly implementing the initial temperature regulation strategy includes calculating the estimated increase in compressor energy consumption based on the target temperature change of the initial temperature regulation strategy. This estimated increase in compressor energy consumption is calculated based on the thermodynamic model of the cigar cabinet and the target temperature change. The estimated increase in compressor energy consumption is then compared with the recent average load represented by the compressor's historical operating load data to calculate the load impact factor. The formula for calculating the load impact factor Ψ is as follows:

[0080]

[0081] in: Indicates the load impact coefficient. This indicates the estimated increase in compressor energy consumption required for the initial temperature control strategy. This represents the recent average load calculated from the compressor's historical workload data. The recent average load is the average energy consumption per unit time of the compressor over the last three hours. The load impact level is determined based on the predefined range of the load impact coefficient Ψ. For example, the predefined range [0, 0.3) corresponds to a low impact level, [0.3, 0.7) corresponds to a medium impact level, and [0.7, +∞) corresponds to a high impact level. In some embodiments, refer to Table 1, the compressor historical workload data table.

[0082] Table 1: Historical Operating Load Data of Compressor

[0083] Timestamp Event Type Duration of a single run (seconds) Average power (watts) 10:05:00 start up 120 105 10:07:00 stop - - 10:25:00 start up 180 110 10:28:00 stop - - 10:50:00 start up 150 108

[0084] It is understandable that the load impact assessment results are used to guide how to decompose the objectives of the initial strategy. When the load impact assessment result is a high impact level, the target temperature change of the initial temperature regulation strategy is decomposed into four or more incremental gradient sub-objectives. For example, the objective of the initial temperature enhancement strategy is a temperature increase of 2.5 degrees Celsius, which is decomposed into five gradient sub-objectives under the high impact level, with each gradient sub-objective increasing the temperature by 0.5 degrees Celsius sequentially. When the load impact assessment result is a medium impact level, the target temperature change of the initial temperature regulation strategy is decomposed into two to three step-like gradient sub-objectives. For example, the same target of a temperature increase of 2.5 degrees Celsius is decomposed into three gradient sub-objectives under the medium impact level, increasing the temperature by 0.8 degrees Celsius, 0.8 degrees Celsius, and 0.9 degrees Celsius respectively. When the load impact assessment result is a low impact level, the target temperature change of the initial temperature regulation strategy is treated as a single gradient sub-objective, i.e., the 2.5-degree Celsius temperature increase is achieved in one step.

[0085] In practical implementation, after decomposing the target of the initial temperature regulation strategy into multiple gradient sub-targets based on the load impact assessment results, the execution timing and intensity parameters are configured for each gradient sub-target to form the final temperature regulation command sequence. The execution timing defines the time point or triggering condition for the start of execution of each gradient sub-target, and the intensity parameter includes the target power or duty cycle setting of the compressor under that gradient sub-target. The final temperature regulation command sequence is an ordered list of commands, and each command in the list includes the action type, target parameters, and execution time point. Data comparison shows that the final temperature regulation command sequence generated by an initial cooling suppression strategy under a high impact assessment contains four compressor frequency reduction operation commands issued at ten-minute intervals with increasing intensity.

[0086] See Figure 4This paper demonstrates the coordinated changes in gradient sub-objective decomposition, execution time, and average compressor power under different load impact levels. At low impact levels, the load impact coefficient Ψ∈[0,0.3), requiring only one gradient sub-objective. The compressor can complete the temperature regulation action in one step, resulting in the shortest total execution time and the highest peak average power (approximately 105 watts), reflecting the direct regulation mode without load buffering. At medium impact levels, the load impact coefficient Ψ∈[0.3,0.7), the strategy objective is decomposed into three stepped gradient sub-objectives, significantly extending the total execution time and reducing the average power to approximately 85 watts, demonstrating the design concept of smoothing load fluctuations through sub-objective decomposition. At high impact levels, the load impact coefficient Ψ∈[0.7,+∞), the objective is decomposed into five incremental gradient sub-objectives, reaching the longest total execution time and further reducing the average power to approximately 65 watts. This mode effectively reduces the instantaneous load impact on the compressor through multi-stage low-intensity regulation, avoiding equipment losses and energy efficiency degradation caused by high-power operation. This parameter correspondence verifies the positive correlation between load impact level and the number of gradient sub-targets, and the negative correlation with average power, providing a quantitative basis for the dynamic load management of cigar cabinet compressors.

[0087] In one embodiment of the present invention, during the execution of monitoring and dynamic adjustment of the command sequence, the original environmental parameter data is continuously collected and updated during the execution of the temperature regulation action. The sensor array works continuously according to the preset sampling frequency. The newly collected temperature data point set, humidity data point set, and pressure data point set are timestamped and then replaced or added to the original original environmental parameter data set, thereby realizing the real-time update of the internal environmental state of the cigar cabinet. Data comparison shows that five minutes after a compressor frequency reduction command is executed, the temperature value recorded by the temperature data point set in the area near the door is 0.3 degrees Celsius lower than before the command is executed.

[0088] In some embodiments, new thermodynamic coupling eigenvalues ​​are calculated in real time based on updated original environmental parameter data. The calculation process is completely consistent with the process of constructing a multidimensional distribution field model and calculating thermodynamic coupling eigenvalues. The difference is that the input data is updated original environmental parameter data obtained in the most recent sampling period. The system periodically calls the eigenvalue calculation engine and outputs a thermodynamic coupling eigenvalue that reflects the latest environmental state. For example, during the execution of the initial cooling suppression strategy, the system calculates a new thermodynamic coupling eigenvalue every minute.

[0089] In practice, the dynamic response rate at which new thermodynamic coupling eigenvalues ​​approach the eigenvalue threshold of a preset optimal cigar maintenance model is monitored. The change in thermodynamic coupling eigenvalues ​​towards the target threshold per unit time is quantified, and its calculation formula is as follows:

[0090]

[0091] in: Indicates the dynamic response rate. Indicates the monitoring time window Within this range, the change in the new thermodynamic coupling eigenvalue relative to the initial eigenvalue points towards the center of the eigenvalue threshold range. This indicates the length of time the monitoring has taken. The calculated dynamic response rate is compared with a preset expected response rate, which is a constant value calibrated based on the thermodynamic characteristics of the cigar cabinet and represents the ideal adjustment speed of the system.

[0092] It is understandable that fine-tuning the execution timing and intensity parameters of subsequent unexecuted gradient sub-objectives based on the dynamic response rate is a feedback control process. If the dynamic response rate is faster than the preset expected response rate, the execution timing of subsequent gradient sub-objectives is appropriately delayed or their intensity parameters are reduced. For example, the issuance time of subsequent compressor start commands is delayed, or the target power setting value of the compressor in subsequent commands is reduced. If the dynamic response rate is slower than the preset expected response rate, the execution timing of subsequent gradient sub-objectives is appropriately advanced or their intensity parameters are increased. For example, the waiting time is shortened, or the operating duty cycle of the compressor in subsequent commands is increased. If the dynamic response rate matches the preset expected response rate, the preset execution timing and intensity parameters in the final temperature regulation command sequence remain unchanged.

[0093] In some embodiments, the fine-tuning operation directly modifies the final temperature regulation instruction sequence stored in memory. The system maintains a queue of instructions to be executed. When fine-tuning is detected, the execution timestamps and intensity parameter values ​​of the instructions that have not yet been executed in the queue are updated. For example, a gradient sub-target originally scheduled to be executed in 10 minutes, with a target temperature reduction of 0.5 degrees Celsius, is modified to be executed in 5 minutes and the target temperature reduction is adjusted to 0.6 degrees Celsius due to a slow dynamic response rate. Optionally, the monitoring time window for the dynamic response rate... This can be a fixed duration, such as five minutes, or a variable duration, such as counting from the start of the previous gradient sub-objective until the current moment. It's understandable that fine-tuning the execution timing and intensity parameters of subsequent unexecuted gradient sub-objectives makes the system's adjustment process more adaptive, avoiding under- or over-adjustment due to model errors or environmental interference. Data comparison shows that in a temperature adjustment process containing three gradient sub-objectives, because the calculated dynamic response rate after the execution of the second gradient sub-objective was 0.02 per minute, faster than the expected response rate of 0.015 per minute, the system delayed the execution of the third gradient sub-objective by eight minutes.

[0094] See Figure 5The heatmap presents the temperature distribution evolution process under different spatial and temporal dimensions. Specifically, the vertical dimension corresponds to four key monitoring areas inside the cigar cabinet: the upper, middle, and lower layers, and the area near the door, representing typical deployment positions of the sensor array in the three-dimensional cavity. The horizontal dimension corresponds to the continuous monitoring time series from 00:00 to 00:18, reflecting the timestamp alignment results of the original environmental parameter data. The heatmap maps temperature values ​​(range 18.5℃-21.5℃) using color coding, which can be directly used as the basic data source for constructing a continuous temperature distribution field. The dramatic fluctuations from dark blue to dark red in the area near the door between 00:02 and 00:04 reflect the transient temperature characteristics caused by heat exchange through the door. This characteristic can be used to identify the airflow static pressure zone and the main heat exchange path when calculating the heat and mass transfer resistance coefficient. The temperature distribution in the upper and middle layers is relatively stable but exhibits gradient differences, providing a direct temperature field gradient vector basis for analyzing the inter-field cooperative change index. The temperature dynamics in the lower layer can be combined with pressure distribution field data to quantitatively assess the impact of airflow on the energy transfer of the temperature field. The multi-temporal and spatial temperature data contained in the heat map is the core input for constructing a multi-dimensional distribution field model of the cigar cabinet and calculating thermodynamic coupling characteristic values, providing a key environmental state characterization basis for the subsequent generation and dynamic adjustment of temperature regulation strategies.

[0095] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A method for intelligent temperature control of a cigar cabinet based on multi-sensor fusion, characterized in that, The method includes: Temperature sensor array, humidity sensor array and airflow pressure sensor array are deployed inside the cigar cabinet to periodically collect raw environmental parameter data of the cigar cabinet's internal cavity; Based on the original environmental parameter data, a multidimensional distribution field model of the internal cavity of the cigar cabinet is constructed, and the thermodynamic coupling characteristic value characterizing the current environmental state is calculated through the multidimensional distribution field model. The thermodynamic coupling eigenvalues ​​are compared with the eigenvalue thresholds of the preset optimal cigar maintenance model, and a preliminary temperature regulation strategy is generated based on the comparison results. Based on the analysis of the preliminary temperature regulation strategy and the historical workload data of the cigar cabinet compressor, a final temperature regulation command sequence containing dynamic execution timing is generated; The final temperature adjustment command sequence is sent to the temperature actuator of the cigar cabinet, driving the temperature actuator to perform the corresponding temperature adjustment action.

2. The intelligent temperature control method for a cigar cabinet based on multi-sensor fusion as described in claim 1, characterized in that, The deployment of temperature sensor arrays, humidity sensor arrays, and airflow pressure sensor arrays inside the cigar cabinet to periodically collect raw environmental parameter data of the cigar cabinet's internal cavity includes: The temperature sensor array, humidity sensor array, and airflow pressure sensor array are respectively arranged in the upper, middle, and lower three layers of the cigar cabinet and in key locations such as near the door. According to the preset sampling frequency, the temperature data point set output by the temperature sensor array, the humidity data point set output by the humidity sensor array, and the pressure data point set output by the airflow pressure sensor array are read and recorded synchronously. The temperature data point set, humidity data point set, and pressure data point set are timestamped and combined to form the original environmental parameter data.

3. The intelligent temperature control method for a cigar cabinet based on multi-sensor fusion as described in claim 2, characterized in that, The construction of a multidimensional distribution field model of the cigar cabinet's internal cavity based on the original environmental parameter data, and the calculation of thermodynamic coupling characteristic values ​​representing the current environmental state using the multidimensional distribution field model, include: The temperature, humidity, and pressure data points, after being aligned with the timestamps, are interpolated and fitted onto the three-dimensional spatial coordinates of the cigar cabinet's internal cavity to generate continuous temperature, humidity, and pressure distribution fields, respectively. Analyze the gradient vector relationship between the continuous temperature distribution field and the continuous humidity distribution field, and calculate the inter-field cooperative change index; Analyze the influence of the continuous pressure distribution field on the energy transfer of the temperature and humidity fields, and calculate the heat and mass transfer resistance coefficient; By integrating the inter-field synergistic change index and the heat and mass transfer resistance coefficient, the thermodynamic coupling characteristic value is output through the characteristic value calculation engine.

4. The intelligent temperature control method for a cigar cabinet based on multi-sensor fusion as described in claim 3, characterized in that, The step of comparing the thermodynamic coupling eigenvalues ​​with the eigenvalue thresholds of a preset optimal cigar maintenance model, and generating a preliminary temperature regulation strategy based on the comparison results, includes: The optimal cigar maintenance model includes a range of feature value thresholds corresponding to an ideal maintenance environment. Determine whether the thermodynamic coupling characteristic value falls within the characteristic value threshold range; If the thermodynamic coupling eigenvalue is lower than the lower limit of the eigenvalue threshold range, a temperature enhancement strategy aimed at improving overall thermodynamic activity is generated. If the thermodynamic coupling eigenvalue is higher than the upper limit of the eigenvalue threshold range, a cooling suppression strategy aimed at reducing overall thermodynamic activity is generated. If the thermodynamic coupling characteristic value falls within the characteristic value threshold range, a isothermal maintenance strategy to maintain the current state is generated. The heating enhancement strategy, cooling suppression strategy, or isothermal maintenance strategy are collectively referred to as the preliminary temperature regulation strategy.

5. The intelligent temperature control method for a cigar cabinet based on multi-sensor fusion as described in claim 1, characterized in that, The analysis based on the preliminary temperature regulation strategy and the historical workload data of the cigar cabinet compressor generates a final temperature regulation command sequence containing dynamic execution timing, including: Extract historical workload data of the compressor within a preset time period from the historical log of the cigar cabinet control unit, including compressor start frequency, single run duration, and power change curve; Analyze the periodic patterns and load intensity trends of the compressor's historical operating load data; Based on the target intensity of the preliminary temperature regulation strategy, assess the potential load impact on the compressor that direct execution of the preliminary temperature regulation strategy may cause; Based on the load impact assessment results, the objectives of the preliminary temperature regulation strategy are decomposed into multiple gradient sub-objectives, and execution timing and intensity parameters are configured for each gradient sub-objective to form the final temperature regulation command sequence.

6. The intelligent temperature control method for a cigar cabinet based on multi-sensor fusion as described in claim 5, characterized in that, The assessment suggests that directly implementing the initial temperature regulation strategy may cause load shocks to the compressor, including: Based on the target temperature change of the preliminary temperature regulation strategy, calculate the required incremental compressor energy consumption. The estimated increase in compressor energy consumption is compared with the recent average load represented in the compressor's historical operating load data to calculate the load impact coefficient. The load impact level is determined based on the predefined range in which the load impact coefficient falls.

7. The intelligent temperature control method for a cigar cabinet based on multi-sensor fusion as described in claim 5, characterized in that, The step of decomposing the objective of the initial temperature regulation strategy into multiple gradient sub-objectives based on the load shock assessment results includes: When the load shock assessment result is a high shock level, the target temperature change of the preliminary temperature regulation strategy is decomposed into four or more incremental gradient sub-objectives. When the load shock assessment result is a medium shock level, the target temperature change of the preliminary temperature regulation strategy is decomposed into two to three step-gradient sub-targets. When the load shock assessment result is a low shock level, the target temperature change of the preliminary temperature regulation strategy is taken as a gradient sub-objective.

8. The intelligent temperature control method for a cigar cabinet based on multi-sensor fusion as described in claim 1, characterized in that, After sending the final temperature adjustment command sequence to the temperature actuator of the cigar cabinet and driving the temperature actuator to perform the corresponding temperature adjustment action, the method further includes: During the execution of the temperature regulation action, the original environmental parameter data are continuously collected and updated; Based on the updated original environmental parameter data, new thermodynamic coupling characteristic values ​​are calculated in real time. Monitor the dynamic response rate at which the new thermodynamic coupling eigenvalue approaches the eigenvalue threshold of the preset optimal cigar maintenance model; Based on the dynamic response rate, the timing and intensity parameters of subsequent unexecuted gradient sub-objectives are fine-tuned.

9. The intelligent temperature control method for a cigar cabinet based on multi-sensor fusion as described in claim 3, characterized in that, The calculation of the heat and mass transfer resistance coefficient includes: Based on the gradient of the continuous pressure distribution field, the main airflow path and static pressure zone of the cigar cabinet's internal cavity are identified; Based on the principles of fluid mechanics, calculate the theoretical mass transfer efficiency along the main path of the airflow. The ratio of the theoretical mass transfer efficiency to the actual mass transfer efficiency derived from changes in the temperature and humidity fields is defined as the heat and mass transfer resistance coefficient.

10. The intelligent temperature control method for a cigar cabinet based on multi-sensor fusion as described in claim 8, characterized in that, The step of fine-tuning the execution timing and intensity parameters of subsequent unexecuted gradient sub-objectives based on the dynamic response rate includes: If the dynamic response rate is faster than the preset expected response rate, the execution timing of subsequent gradient sub-objectives will be appropriately delayed or their intensity parameters will be reduced. If the dynamic response rate is slower than the preset expected response rate, the execution timing of subsequent gradient sub-objectives should be advanced or their intensity parameters should be increased. If the dynamic response rate matches the preset expected response rate, then the preset execution timing and intensity parameters in the final temperature adjustment command sequence remain unchanged.

Images