Temperature monitoring method and system for Chinese mugwort active ingredient storage equipment

Abstract

The invention belongs to the technical field of temperature monitoring, particularly relates to a temperature monitoring method and system for Chinese mugwort active ingredient storage equipment, and solves the problem that the temperature in the Chinese mugwort storage equipment is difficult to intelligently monitor, regulate and control by the existing temperature control technology. The monitoring method comprises the following steps: S1, carrying out weighted fusion on real-time temperature data of a plurality of sensors, and calculating to obtain an in-cabin reference temperature at the current moment; s2, calculating an equivalent temperature integral value for representing the accumulated loss risk of the active component; correcting a spatial thermal field distribution weight coefficient based on the equivalent temperature integral value; s3, selecting or configuring a corresponding prediction model, and resolving a temperature prediction value in a future short-time domain; and S4, generating and outputting a temperature regulation and control instruction. According to the method, the temperature change is pre-judged and intervened in advance, passive response type control is changed into active preventive regulation, and the medicinal value of Chinese mugwort is guaranteed.

Description

Technical Field

[0001] This invention belongs to the technical field of temperature monitoring, specifically relating to a temperature monitoring method and system for a device for preserving the active ingredients of Artemisia argyi. Background Technology

[0002] As a national geographical indication product, the medicinal and economic value of Qi Ai (Artemisia argyi) highly depends on its active ingredients. These active ingredients are extremely sensitive to environmental temperature during storage, transportation, and processing. Sustained high temperatures or drastic temperature fluctuations accelerate the loss and decomposition of key components, leading to decreased efficacy and quality degradation, thus impacting the profitability of the entire industry chain. Therefore, how to intelligently monitor and control the internal temperature of Qi Ai storage equipment to effectively mitigate the loss of active ingredients has become a critical technical challenge that urgently needs to be addressed in the current industrialization of Qi Ai.

[0003] Existing storage or preservation equipment generally uses fixed-point temperature acquisition, comparing the measured temperature with preset constant upper and lower limits. Once the temperature exceeds the limit, cooling or heating devices are activated. However, this method ignores the non-uniformity of the temperature field distribution inside the equipment. Simple multi-point averages or extreme values ​​cannot accurately reflect the overall thermal field state inside the chamber, which may lead to the failure to detect local overheating or undercooling areas in a timely manner. At the same time, its control threshold remains unchanged and cannot be dynamically adjusted according to the risk of accumulated loss of active ingredients. In addition, the control logic lacks predictive ability and can only respond passively after temperature anomalies occur. It cannot predict temperature change trends in advance and take preventive measures, resulting in significant control lag, which is difficult to meet the preservation requirements of high-value and highly sensitive materials such as Artemisia argyi. Summary of the Invention

[0004] This invention provides a temperature monitoring method and system for a device for preserving the active ingredients of Artemisia argyi, solving the problem that existing temperature control technologies are unable to intelligently monitor and regulate the internal temperature of Artemisia argyi preservation devices.

[0005] In a first aspect, the present invention provides a temperature monitoring method for a device for preserving the active ingredients of Artemisia argyi, comprising the following steps: S1. Acquire real-time temperature data from multiple preset sensors within the storage device; based on a set of spatial thermal field distribution weighting coefficients, perform weighted fusion of the real-time temperature data from multiple sensors to calculate the current cabin reference temperature. S2, based on the historical cabin reference temperature sequence updated to the current moment and the preset nonlinear dissipation characteristic function between the volatile active ingredients of Artemisia argyi and temperature, calculate the equivalent temperature integral value used to characterize the cumulative loss risk of active ingredients; based on the equivalent temperature integral value, correct the weighting coefficient of the spatial thermal field distribution. S3. Based on the short-term fluctuation rate of the cabin reference temperature and the equivalent temperature integral value, select or configure the corresponding prediction model to calculate the predicted temperature value in the short time domain in the future. S4. When the cabin reference temperature exceeds the safety threshold range adjusted based on the equivalent temperature integral value, or when the equivalent temperature integral value is greater than the cumulative risk threshold, a temperature control command is generated and output.

[0006] Furthermore, using the formula Calculate the cabin reference temperature, wherein, The cabin reference temperature. The spatial thermal field distribution weighting coefficient for the i-th sensor is... Let represent the real-time temperature data of the i-th sensor, and N be the total number of sensors.

[0007] Furthermore, the discrete integral formula is adopted. Calculate the equivalent temperature integral value, where, This is the equivalent temperature integral value at the current time k. Let A be the equivalent temperature integral value from the previous moment, and A be the pre-exponential factor. Let R be the activation energy for dissipation, and R be the ideal gas constant. This represents the absolute temperature value corresponding to the current reference temperature inside the cabin. For time step.

[0008] Furthermore, based on the equivalent temperature integral value, the weighting coefficient of the spatial thermal field distribution is corrected, including the following steps: According to the revised formula Calculate the initial correction weights for each sensor. ; For all initial adjusted weights After normalization, the updated spatial thermal field distribution weighting coefficients are obtained. ,in, The preset dimensionless correction coefficient, Let be the historical average absolute temperature of the i-th sensor. The historical average baseline absolute temperature is given, and ETI is the current equivalent temperature integral value. The cumulative risk threshold is N, the total number of sensors is N, and j is the index variable.

[0009] Furthermore, based on the short-term fluctuation rate of the cabin reference temperature and the equivalent temperature integral value, a corresponding prediction model is selected or configured to calculate the predicted temperature value in the short-term time domain, including the following steps: The autoregressive integral moving average model ARIMA(p, d, q) is used as the prediction model, where the preset model is dynamically configured according to the following rules: The parameter p is set as follows: when the short-term volatility is greater than 0.05℃ / minute in the past 30 minutes, p=5; otherwise, p=3. The parameter d is set as follows: when the equivalent temperature integral value exceeds 75% of the cumulative risk threshold, d=2; otherwise, d=1. The parameter q is set to 2.

[0010] Furthermore, when the cabin reference temperature exceeds the safety threshold range adjusted based on the equivalent temperature integral value, or when the equivalent temperature integral value is greater than the cumulative risk threshold, a temperature control command is generated and output, including the following steps: Set the safety threshold range as follows ,in: lower limit ; upper limit ; in, To maintain the target temperature, The base temperature deviation is ETI, which is the integral value of the current equivalent temperature. This is the cumulative risk threshold; When the cabin reference temperature is lower than or higher At that time, heating or cooling commands are generated respectively.

[0011] Secondly, the present invention provides a temperature monitoring system for a device for preserving the active ingredients of Artemisia argyi, comprising: The calculation module acquires real-time temperature data from multiple preset sensors stored in the device; based on a set of spatial thermal field distribution weighting coefficients, it performs weighted fusion of the real-time temperature data from multiple sensors to calculate the current cabin reference temperature. The correction module calculates the equivalent temperature integral value to characterize the cumulative loss risk of active ingredients based on the historical cabin reference temperature sequence updated to the current time and the preset nonlinear dissipation characteristic function between the volatile active ingredients of Artemisia argyi and temperature; and corrects the spatial thermal field distribution weighting coefficient based on the equivalent temperature integral value. The calculation module selects or configures the corresponding prediction model based on the short-term fluctuation rate of the cabin reference temperature and the equivalent temperature integral value, and calculates the predicted temperature value in the short time domain in the future. The generation module generates and outputs a temperature control command when the cabin reference temperature exceeds the safety threshold range adjusted based on the equivalent temperature integral value, or when the equivalent temperature integral value is greater than the cumulative risk threshold.

[0012] Furthermore, using the formula Calculate the cabin reference temperature, wherein, The cabin reference temperature. The spatial thermal field distribution weighting coefficient for the i-th sensor is... Let represent the real-time temperature data of the i-th sensor, and N be the total number of sensors.

[0013] Furthermore, the discrete integral formula is adopted. Calculate the equivalent temperature integral value, where, This is the equivalent temperature integral value at the current time k. Let A be the equivalent temperature integral value from the previous moment, and A be the pre-exponential factor. Let R be the activation energy for dissipation, and R be the ideal gas constant. This represents the absolute temperature value corresponding to the current reference temperature inside the cabin. For time step.

[0014] Furthermore, based on the equivalent temperature integral value, the weighting coefficient of the spatial thermal field distribution is corrected, including the following steps: According to the revised formula Calculate the initial correction weights for each sensor. ; For all initial adjusted weights After normalization, the updated spatial thermal field distribution weighting coefficients are obtained. ,in, The preset dimensionless correction coefficient, Let be the historical average absolute temperature of the i-th sensor. The historical average baseline absolute temperature is given, and ETI is the current equivalent temperature integral value. The cumulative risk threshold is N, the total number of sensors is N, and j is the index variable.

[0015] The beneficial effects are as follows: Compared with existing technologies, this invention obtains a cabin reference temperature that comprehensively represents the overall thermal field distribution inside the equipment by weighted and fused real-time temperature data from multiple sensors, overcoming the monitoring inaccuracy problem caused by uneven temperature distribution in traditional methods. This invention transforms the goal of temperature control from maintaining a constant temperature range to managing and controlling the chronic loss process of active ingredients by using an equivalent temperature integral value representing the cumulative loss risk of Artemisia argyi. By correlating the safety control threshold with this equivalent temperature integral value, the temperature control strategy can be adjusted according to the degree of quality loss that has occurred, and strict control can be implemented when the cumulative risk is high, thereby achieving deep protection of the quality of Artemisia argyi. Furthermore, by using a predictive model to calculate future temperatures, it achieves early prediction and intervention of temperature changes, transforming passive responsive control into proactive preventative regulation, thus ensuring the medicinal value of Artemisia argyi. Attached Figure Description

[0016] Figure 1A flowchart of a temperature monitoring method for equipment used to preserve the active ingredients of Artemisia argyi; Figure 2 A schematic diagram showing the integral values ​​of the reference temperature and equivalent temperature inside the cabin; Figure 3 A schematic diagram of the selection mechanism for the prediction model; Figure 4 This is a schematic diagram of the safety threshold range. Detailed Implementation

[0017] An embodiment of the temperature monitoring method for the Artemisia argyi active ingredient preservation device provided by the present invention: like Figure 1 As shown, a temperature monitoring method for a device for preserving active ingredients of Artemisia argyi includes the following steps: S1. Acquire real-time temperature data from multiple preset sensors within the storage device; based on a set of spatial thermal field distribution weighting coefficients, perform weighted fusion of the real-time temperature data from multiple sensors to calculate the current cabin reference temperature. Inside the Artemisia argyi preservation device, multiple DS18B20 digital temperature sensors are arranged in several key locations, such as the top, middle, bottom, and near the door and refrigeration unit. A microcontroller, such as an STM32 series microcontroller, polls all sensors via a single-bus protocol, reads and acquires the current temperature value of each sensor node, and obtains a real-time dataset containing the temperature of all measuring points.

[0018] A weighting coefficient is pre-set for each sensor location. For example, the weighting coefficient for the sensor in the central area is set to 0.4, the weighting coefficients for the top and bottom sensors are each set to 0.2, and the weighting coefficients for the sensors near the door and the refrigeration unit are each set to 0.1. The sum of all weighting coefficients is 1. The real-time temperature value of each sensor obtained in the first step is multiplied by its corresponding weighting coefficient, and then all the products are added together. The result is the current cabin reference temperature.

[0019] In an optional embodiment, the formula is used. Calculate the cabin reference temperature, wherein, The cabin reference temperature. The spatial thermal field distribution weighting coefficient for the i-th sensor is... Let represent the real-time temperature data of the i-th sensor, and N be the total number of sensors.

[0020] For example, four temperature sensors, N = 4, are installed in a smart moxibustion storage cabinet. Different weighting coefficients are pre-set based on their location and representativeness of the overall ambient temperature. Sensor 1, located near the door, is more susceptible to external interference, so its weight is lower. Set to 0.1; Sensor 2, located in the center of the cabin, reflects the main storage environment, weight... Set to 0.4; 3 weights for sensors near the cooling unit. Set to 0.3; Sensor 4 weight at the top Set it to 0.2.

[0021] At a certain moment, the real-time temperature data collected from the four sensors were as follows: It is 22.5℃. It is 21.0℃. It is 20.5℃. The temperature is 21.8℃. Using this formula, the cabin reference temperature is calculated to be 21.16℃. This result incorporates the temperature contributions from different regions.

[0022] S2, based on the historical cabin reference temperature sequence updated to the current moment and the preset nonlinear dissipation characteristic function between the volatile active ingredients of Artemisia argyi and temperature, calculate the equivalent temperature integral value used to characterize the cumulative loss risk of active ingredients; based on the equivalent temperature integral value, correct the weighting coefficient of the spatial thermal field distribution. Specifically, in one embodiment, the nonlinear dissipation characteristic function is a loss rate model based on the Arrhenius equation, mapping the cabin reference temperature to the instantaneous loss rate of a volatile component. The cabin reference temperature is continuously recorded, forming a time series. In each calculation cycle, e.g., every minute, the average reference temperature within that cycle is taken, substituted into the aforementioned function to calculate the average loss rate for that minute, and then multiplied by the time interval of one minute to obtain the loss increment for that minute. This loss increment is accumulated to the cumulative loss value of the previous moment to obtain the equivalent temperature integral value updated to the current moment, such as... Figure 2 .

[0023] A correction trigger threshold is set. When the equivalent temperature integral value exceeds this threshold, historical data is analyzed to identify sensors with consistently high readings over a past period. The weighting coefficients of these sensors are slightly increased, for example, by 0.02, while the weighting coefficients of other sensors, especially those with consistently low readings, are proportionally decreased, ensuring that the sum of all weighting coefficients remains 1. The corrected weighting coefficients will be used in the next weighted fusion calculation.

[0024] In an optional embodiment, a discrete integral formula is used. Calculate the equivalent temperature integral value, where, This is the equivalent temperature integral value at the current time k. Let A be the equivalent temperature integral value from the previous moment, and A be the pre-exponential factor. Let R be the activation energy for dissipation, and R be the ideal gas constant. This represents the absolute temperature value corresponding to the current reference temperature inside the cabin. For time step.

[0025] Discrete temperature data points are transformed into a continuously accumulated index to assess the degree of loss of volatile components in Artemisia argyi due to heat. The above formula is the discrete integral form of the Arrhenius equation, representing the nonlinear relationship between the chemical reaction rate and temperature. An activation energy for dissipation in Artemisia argyi is assumed. It is 83140 J / mol, and the pre-exponential factor A is / s, the ideal gas constant R is 8.314 J / (mol·K), and the time step Δt is 60 s.

[0026] At time k-1, the calculated equivalent temperature integral value ETI is 50.0. At the current time k, the measured cabin reference temperature is 21.16℃, which, converted to absolute temperature... That is, 294.31 K. Substituting these data into the formula, the newly added equivalent temperature integral value within the current time step is calculated, which is approximately 0.015. Therefore, the cumulative equivalent temperature integral value at the current time k is... Updated to 50.015. This reflects the cumulative risk of heat damage to the quality of Qi Ai (Artemisia argyi) to date.

[0027] In an optional embodiment, the spatial thermal field distribution weighting coefficient is corrected based on the equivalent temperature integral value, including the following steps: According to the revised formula Calculate the initial correction weights for each sensor. ; For all initial adjusted weights After normalization, the updated spatial thermal field distribution weighting coefficients are obtained. ,in, The preset dimensionless correction coefficient, Let be the historical average absolute temperature of the i-th sensor. The historical average baseline absolute temperature is given, and ETI is the current equivalent temperature integral value. The cumulative risk threshold is N, the total number of sensors is N, and j is the index variable.

[0028] When the equivalent temperature integral (ETI) increases, attention should be paid to regions that have historically been consistently hotter. For example, a cumulative risk threshold can be set. The value is 1000, and the correction factor β is 0.5. The current ETI has reached 800, and the historical average absolute temperature is 293K. Sensor 1 has been located in a relatively hot region for a long time, and its historical average absolute temperature is 294K. Its current weight is 0.1.

[0029] The initial corrected weight for sensor 1 is calculated to be approximately 0.10014. Conversely, sensor 3 is located in a persistently cold region with a historical average temperature of 292K, and its current weight is 0.3; its initial corrected weight is calculated to be approximately 0.29959. After normalizing the initial corrected weights for all sensors, the new weight for sensor 1 will increase slightly, while the new weight for sensor 3 will decrease slightly.

[0030] S3. Based on the short-term fluctuation rate of the cabin reference temperature and the equivalent temperature integral value, select or configure the corresponding prediction model to calculate the predicted temperature value in the short time domain in the future. Specifically, the standard deviation of the cabin baseline temperature over the past 15 minutes is calculated as the short-term volatility. When volatility is low and the equivalent temperature integral is at a safe level, the Autoregressive Moving Average (ARIMA) model is used for prediction. When volatility increases or the equivalent temperature integral approaches a warning level, the system switches to a pre-trained Long Short-Term Memory (LSTM) network model, such as... Figure 3 This model can identify nonlinear trends and thus calculate temperature changes over the next 5 minutes.

[0031] In an optional embodiment, based on the short-term volatility of the cabin reference temperature and the equivalent temperature integral value, a corresponding prediction model is selected or configured to calculate the predicted temperature value in the short-term time domain, including the following steps: The autoregressive integral moving average model ARIMA(p, d, q) is used as the prediction model, where the preset model is dynamically configured according to the following rules: The parameter p is set as follows: when the short-term volatility is greater than 0.05℃ / minute in the past 30 minutes, p=5; otherwise, p=3. The parameter d is set as follows: when the equivalent temperature integral value exceeds 75% of the cumulative risk threshold, d=2; otherwise, d=1. The parameter q is set to 2.

[0032] By configuring the parameters p and d of the ARIMA model, optimized predictions for different operating conditions can be achieved. For example, in a stable operating phase, the cabin baseline temperature has changed gradually over the past 30 minutes, with a short-term volatility of 0.03°C per minute, below the threshold of 0.05. Meanwhile, the current equivalent temperature integral value (ETI) is 600, while the risk threshold is 1000, which is less than 75% of the threshold.

[0033] The model parameters p are configured as 3, d as 1, and q is fixed at 2, employing an ARIMA third-order first-order differencing second-order moving average model for forecasting. This model is suitable for stationary or simple trend time series. However, if frequent door opening and closing causes drastic temperature fluctuations, with the volatility rising to 0.08℃ per minute and the ETI accumulating to 800, exceeding the 75% threshold, parameter p is adjusted to 5 to capture more recent temperature change history, while d is adjusted to 2 to perform second-order differencing to address strong non-stationarity in the data. The model then becomes an ARIMA fifth-order second-order differencing second-order moving average model.

[0034] S4. When the cabin reference temperature exceeds the safety threshold range adjusted based on the equivalent temperature integral value, or when the equivalent temperature integral value is greater than the cumulative risk threshold, a temperature control command is generated and output.

[0035] Specifically, the initial first safety threshold range is set to 4°C to 10°C. The upper limit of this range decreases linearly with the increase of the equivalent temperature integral value; for example, the upper temperature limit decreases by 0.5°C for every 10-unit increase in the integral value. When the real-time cabin reference temperature is lower than the adjusted lower limit or higher than the adjusted upper limit, a control signal is output to the relay of the cooling or heating unit to start or stop the corresponding equipment. At the same time, a fixed cumulative risk threshold is set; if the equivalent temperature integral value exceeds this value, forced cooling is activated, and a remote alarm message is sent through the network module.

[0036] In an optional embodiment, when the cabin reference temperature exceeds the safety threshold range adjusted based on the equivalent temperature integral value, or when the equivalent temperature integral value is greater than the cumulative risk threshold, a temperature control command is generated and output, including the following steps: Set the safety threshold range as follows ,in: lower limit ; upper limit ; in, To maintain the target temperature, The base temperature deviation is ETI, which is the integral value of the current equivalent temperature. This is the cumulative risk threshold; When the cabin reference temperature is lower than or higher At that time, heating or cooling commands are generated respectively.

[0037] Specifically, this embodiment establishes a tightened temperature control range, making the control strategy more stringent as the accumulated risk of the product increases. Assuming a target storage temperature... The base temperature is 20℃, the baseline temperature deviation ΔT is 2℃, and the cumulative risk threshold is... The value is 1000. In the initial storage phase, the current equivalent temperature integral (ETI) is 100. At this point, the calculated safe lower limit is 18.2℃, and the upper limit is 21.8℃. The safe threshold range at this time is 18.2-21.8℃, a relatively wide range that allows for some temperature fluctuations. Figure 4 .

[0038] As storage time increased, the ETI value rose to 900, approaching the risk threshold. The safety threshold range was recalculated using the same formula. The lower limit became 19.8℃, and the upper limit became 20.2℃. The safety threshold range narrowed to 19.8 to 20.2℃. Temperature control became more sensitive; if the reference temperature deviated from the target value by more than 0.2℃, a cooling or heating command was generated. This ensured that further deterioration was mitigated through temperature control when product quality risks accumulated to a high level.

[0039] An embodiment of the temperature monitoring system for the Artemisia argyi active ingredient preservation device provided by the present invention: The temperature monitoring system for the preservation equipment of Artemisia argyi active ingredients includes: The calculation module acquires real-time temperature data from multiple preset sensors stored in the device; based on a set of spatial thermal field distribution weighting coefficients, it performs weighted fusion of the real-time temperature data from multiple sensors to calculate the current cabin reference temperature. The correction module calculates the equivalent temperature integral value to characterize the cumulative loss risk of active ingredients based on the historical cabin reference temperature sequence updated to the current time and the preset nonlinear dissipation characteristic function between the volatile active ingredients of Artemisia argyi and temperature; and corrects the spatial thermal field distribution weighting coefficient based on the equivalent temperature integral value. The calculation module selects or configures the corresponding prediction model based on the short-term fluctuation rate of the cabin reference temperature and the equivalent temperature integral value, and calculates the predicted temperature value in the short time domain in the future. The generation module generates and outputs a temperature control command when the cabin reference temperature exceeds the safety threshold range adjusted based on the equivalent temperature integral value, or when the equivalent temperature integral value is greater than the cumulative risk threshold.

[0040] Furthermore, using the formula Calculate the cabin reference temperature, wherein, The cabin reference temperature. The spatial thermal field distribution weighting coefficient for the i-th sensor is... Let represent the real-time temperature data of the i-th sensor, and N be the total number of sensors.

[0041] Furthermore, the discrete integral formula is adopted. Calculate the equivalent temperature integral value, where, This is the equivalent temperature integral value at the current time k. Let A be the equivalent temperature integral value from the previous moment, and A be the pre-exponential factor. Let R be the activation energy for dissipation, and R be the ideal gas constant. This represents the absolute temperature value corresponding to the current reference temperature inside the cabin. For time step.

[0042] Furthermore, based on the equivalent temperature integral value, the weighting coefficient of the spatial thermal field distribution is corrected, including the following steps: According to the revised formula Calculate the initial correction weights for each sensor. ; For all initial adjusted weights After normalization, the updated spatial thermal field distribution weighting coefficients are obtained. ,in, The preset dimensionless correction coefficient, Let be the historical average absolute temperature of the i-th sensor. The historical average baseline absolute temperature is given, and ETI is the current equivalent temperature integral value. The cumulative risk threshold is N, the total number of sensors is N, and j is the index variable.

[0043] In addition, in the description of this specification, "multiple" means at least two, such as two, three or more, etc., unless otherwise expressly and specifically defined.

Claims

1. A temperature monitoring method of a Wucai active ingredient preservation device, characterized in that, The method comprises the following steps: S1, obtaining real-time temperature data of a plurality of sensors preset in the device; Based on a set of spatial thermal field distribution weight coefficients, the real-time temperature data of the plurality of sensors is weighted and fused to calculate the cabin reference temperature at the current time; S2, according to the historical cabin reference temperature sequence updated to the current time and the preset nonlinear dispersion characteristic function between the volatile active ingredients of Radix Wuligan and temperature, the equivalent temperature integral value for representing the cumulative loss risk of active ingredients is calculated; based on the equivalent temperature integral value, the spatial thermal field distribution weight coefficient is corrected; S3, according to the short-term fluctuation rate of the cabin reference temperature and the equivalent temperature integral value, a corresponding prediction model is selected or configured to calculate the temperature prediction value in the future short time domain; S4, when the cabin reference temperature exceeds the safety threshold range adjusted based on the equivalent temperature integral value, or the equivalent temperature integral value is greater than the cumulative risk threshold, a temperature control instruction is generated and output.

2. The temperature monitoring method of the Wucai active ingredient preservation device according to claim 1, characterized in that, The cabin reference temperature is calculated by the formula wherein, is the cabin reference temperature, is the spatial heat field distribution weight coefficient of the ith sensor, is the real-time temperature data of the ith sensor, and N is the total number of sensors.

3. The temperature monitoring method of the Wucai active ingredient preservation device according to claim 1, characterized in that, using a discrete integration formula calculating the equivalent temperature integral value, wherein is the equivalent temperature integral value at the current time k, is the equivalent temperature integral value at the previous time, and A is a previous factor, is the activation energy of the evaporation, and R is the ideal gas constant, is the absolute temperature value corresponding to the cabin reference temperature at the current time, is the time step.

4. The temperature monitoring method of the Wucai active ingredient preservation device according to claim 3, characterized in that, Based on the equivalent temperature integral value, the spatial thermal field distribution weight coefficient is corrected, which comprises the following steps: According to the correction formula Calculating the preliminary correction weight of each sensor ; for all preliminary correction weights normalization to obtain updated spatial heat field distribution weight coefficients wherein, is a preset dimensionless correction coefficient, is a historical average absolute temperature of the ith sensor, is a historical average reference absolute temperature, ETI is a current equivalent temperature integral value, is a cumulative risk threshold value, N is a total number of sensors, and j is an index variable.

5. The temperature monitoring method of the Wucai active ingredient preservation device according to claim 4, characterized in that, According to the short-term fluctuation rate of the cabin reference temperature and the equivalent temperature integral value, a corresponding prediction model is selected or configured to calculate the temperature prediction value in the future short time domain, which comprises the following steps: An autoregressive integrated moving average model ARIMA (p, d, q) is used as the prediction model, wherein the preset model is dynamically configured according to the following rules: The setting mode of the parameter p is: when the short-term fluctuation rate is greater than 0.05℃ / minute within the past 30 minutes, p=5, otherwise p=3; The setting mode of the parameter d is: when the equivalent temperature integral value exceeds 75% of the cumulative risk threshold, d=2, otherwise d=1; The parameter q is set to 2.

6. The temperature monitoring method of the Wucai active ingredient preservation device according to claim 4, characterized in that, When the cabin reference temperature exceeds the safety threshold range adjusted based on the equivalent temperature integral value, or the equivalent temperature integral value is greater than the cumulative risk threshold, a temperature control instruction is generated and output, which comprises the following steps: Setting the safety threshold range to wherein: lower limit ; upper limit ; wherein, Ttarget is the target holding temperature, Tbase is a base temperature offset, ETI is the current equivalent temperature integral, is a cumulative risk threshold; When the cabin reference temperature is lower than or higher than , heating or refrigeration instructions are generated, respectively.

7. A temperature monitoring system for a Wucai active ingredient storage device, characterized in that, It comprises: A calculation module obtains real-time temperature data of a plurality of sensors preset in the device; Based on a set of spatial thermal field distribution weight coefficients, the real-time temperature data of the plurality of sensors is weighted and fused to calculate the cabin reference temperature at the current time; A correction module calculates an equivalent temperature integral value for representing the cumulative loss risk of active ingredients according to the historical cabin reference temperature sequence updated to the current time and the preset nonlinear dispersion characteristic function between the volatile active ingredients of Radix Wuligan and temperature; based on the equivalent temperature integral value, the spatial thermal field distribution weight coefficient is corrected; A calculation module, according to the short-term fluctuation rate of the cabin reference temperature and the equivalent temperature integral value, selects or configures a corresponding prediction model to calculate the temperature prediction value in the future short time domain; A generation module generates and outputs a temperature control instruction when the cabin reference temperature exceeds the safety threshold range adjusted based on the equivalent temperature integral value, or the equivalent temperature integral value is greater than the cumulative risk threshold.

8. The temperature monitoring system of the Wucai active ingredient preservation device according to claim 7, characterized in that, The cabin reference temperature is calculated by the formula wherein, is the cabin reference temperature, is the spatial thermal field distribution weight coefficient of the ith sensor, is the real-time temperature data of the ith sensor, and N is the total number of sensors.

9. The temperature monitoring system of the Wucai active ingredient preservation device according to claim 7, characterized in that, using a discrete integration formula calculating the equivalent temperature integral value, wherein is the equivalent temperature integral value at the current time k, is the equivalent temperature integral value at the previous time, A is a is the activation energy of the evaporation, R is the ideal gas constant, is the absolute temperature value corresponding to the cabin reference temperature at the current time, is the time step.

10. The temperature monitoring system of the Wucai active ingredient preservation device according to claim 9, characterized in that, Based on the equivalent temperature integral value, the spatial thermal field distribution weight coefficient is corrected, which comprises the following steps: According to the correction formula Calculating the preliminary correction weight of each sensor ; all preliminary correction weights normalization to obtain updated spatial heat field distribution weight coefficients wherein, is a preset dimensionless correction coefficient, is a historical average absolute temperature of the i-th sensor, is a historical average reference absolute temperature, ETI is a current equivalent temperature integral value, is a cumulative risk threshold value, N is the total number of sensors, and j is an index variable.

Images