Temperature monitoring method and system for vaporizer

Abstract

The invention relates to the technical field of data processing and analysis, in particular to a temperature monitoring method and system for a # imgabs0 # vaporizer. The method comprises the following steps: calculating an optimal coefficient; according to the optimal coefficient, using an exponential smoothing method to predict the temperature; the step of calculating the optimal coefficient comprises the following steps: acquiring temperature data at a historical moment, and selecting a preset number of data to construct an adjacent data set at the current moment; calculating a local fluctuation degree and a trend influence factor at the current moment according to the temperature data in the adjacent data set, and taking a product of the local fluctuation degree and the trend influence factor as an adjustment coefficient; and setting an initial smoothing coefficient, and adjusting the initial smoothing coefficient through the adjusting coefficient to obtain an optimal coefficient. The method has the effect of improving the reaction capacity of the vaporizer to the real-time temperature change in the temperature prediction process.

Description

Technical Field

[0001] The present application relates to the field of data processing and analysis technology, and in particular to a Temperature monitoring method and system for a vaporizer. Background Art

[0002] Vaporized hydrogen peroxide sterilization is a sterilization method that converts liquid hydrogen peroxide into gaseous hydrogen peroxide at a certain temperature. It has the advantages of being dry, fast-acting, non-toxic and residue-free, and is used in biotechnology, medicine and health industries. In this process, liquid hydrogen peroxide is mainly heated by a vaporizer to convert it into a gaseous state, and then diffused into the sterilization area to kill microorganisms. The temperature control during the heating and vaporization process is more sensitive. If the vaporizer temperature is low and the hydrogen peroxide vaporization rate is slow, the efficiency of disinfection and sterilization will be low. When the temperature of the vaporizer is too high, hydrogen peroxide will accelerate the decomposition into water and oxygen, resulting in a decrease in the concentration of hydrogen peroxide in the sterilization area, affecting the sterilization efficiency, or causing an increase in equipment pressure and causing a safety accident. Therefore, monitoring the temperature during the vaporization process of the vaporizer and early warning are important protective measures to reduce safety risks.

[0003] Exponential smoothing is a commonly used method in production forecasting. It can be used with a certain time series model to predict the future changes in vaporizer temperature. Its main principle is that the exponential smoothing value of any period is the weighted average of the actual observation value of the current period and the exponential smoothing value of the previous period. However, during the process of vaporizing hydrogen peroxide in the vaporizer, its temperature curve is not completely smooth, but has certain fluctuations. The traditional exponential smoothing method has a certain lag in prediction. The carburetor has a poor ability to respond to real-time temperature changes. Summary of the invention

[0004] In order to solve the traditional prediction method To solve the problem that the carburetor has a weak ability to respond to real-time temperature changes, the present application provides a Temperature monitoring method and system for a vaporizer.

[0005] In the first aspect, the present application provides a The temperature monitoring method for the vaporizer adopts the following technical solution: A sort of A temperature monitoring method for a vaporizer includes the steps of: calculating an optimal coefficient; predicting the temperature using an exponential smoothing method according to the optimal coefficient; The steps of calculating the optimal coefficient include: obtaining temperature data at historical moments, and selecting a preset number of data to construct a current near-data set; calculating the local fluctuation degree and trend influence factor at the current moment according to the temperature data in the near-data set; taking the product of the local fluctuation degree and the trend influence factor as the adjustment coefficient; setting an initial smoothing coefficient, and adjusting the initial smoothing coefficient by the adjustment coefficient to obtain the optimal coefficient; wherein, the calculation formula of the trend influence factor is: ; Where: Indicates The trend influencing factor of temperature data is Indicates the number of temperature data in the adjacent data set. Indicates The temperature data in the adjacent data set Temperature measurements, Indicates The temperature data in the adjacent data set The predicted temperature value, Indicates The temperature data is close to the extreme value of the data set. Indicates The weighted average of the temperature data sets adjacent to the data set.

[0006] In this application, on the one hand, the changing pattern of the temperature data in the adjacent data set at the current moment is taken into account. The smoothing coefficient used in the exponential smoothing prediction is adjusted according to the changing pattern of the adjacent data set. For example, when the temperature in the adjacent data set changes over time (the temperature gradually increases or decreases), the initial smoothing coefficient is adjusted to increase it, so that more attention can be paid to the adjacent data, reducing the dependence on the previous prediction data, and improving the prediction model's ability to respond to real-time temperature changes. On the other hand, in the formula The difference between the previous predicted values ​​and the measured values ​​is partially taken into account, so as to reflect the prediction accuracy. The smoothing coefficient of the exponential smoothing method prediction is adjusted according to the prediction accuracy. For example, when the difference between the predicted value and the measured value is large, the initial smoothing coefficient is increased to pay more attention to the adjacent data, and vice versa, so as to improve the prediction accuracy.

[0007] Optionally, the weighted average of adjacent data sets is calculated as: ; In the formula, Indicates The weighted average of the adjacent data sets of temperature data, Indicates the number of data in the adjacent data set. Indicates The temperature data in the adjacent data set A temperature measurement.

[0008] Optionally, the local fluctuation degree is calculated as: ; In the formula, Indicates The weighted average of the temperature data sets adjacent to the data set. Indicates the number of data in the adjacent data set. Indicates The temperature data in the adjacent data set Temperature measurements; Indicates The local fluctuation degree of temperature data, represents the maximum value function; is an exponential function with a natural constant as its base.

[0009] Calculate the difference between the temperature data in the adjacent data set and the overall weighted average to reflect the fluctuation range between the temperatures within the adjacent data set. When the fluctuation range is large, more attention should be paid to the recent data to improve accuracy.

[0010] Optionally, the local fluctuation degree is calculated as: ; In the formula, Indicates The weighted average of the temperature data sets adjacent to the data set. Indicates the number of data in the adjacent data set. Indicates The temperature data in the adjacent data set Temperature measurements, Indicates The local fluctuation degree of temperature data, represents the maximum value function; is an exponential function with a natural constant as its base.

[0011] Guaranteed by absolute value operation The value of is a positive number to avoid affecting subsequent calculations. At the same time, the absolute value calculation is simpler and improves the calculation efficiency. In some scenarios with high requirements for calculation efficiency, this method can ensure the prediction accuracy while taking into account the calculation efficiency.

[0012] Optionally, the step of predicting the temperature using an exponential smoothing method according to the optimal coefficient includes: monitoring the vaporizer temperature according to the result of the temperature prediction, and alarming when the temperature change of the vaporizer is inconsistent with the operating state of the vaporizer.

[0013] The temperature of the vaporizer is monitored according to the results of temperature prediction. When the vaporizer starts to be used, the temperature of the vaporizer should gradually increase. Under the current state, if the predicted results show that the temperature of the vaporizer is stable or gradually decreases, it means that the vaporizer is faulty.

[0014] Optionally, the step of adjusting the initial smoothing coefficient to obtain the optimal coefficient includes: obtaining the adjustment coefficients of multiple temperature data in historical moments and calculating the average value; obtaining the adjustment coefficient at the current moment and calculating the ratio between it and the average value, and taking the product of the initial smoothing coefficient and the ratio as the optimal coefficient.

[0015] Optionally, the step of adjusting the initial smoothing coefficient to obtain the optimal coefficient includes: obtaining the adjustment coefficient of the temperature data with a time interval of less than 20 minutes from the current moment, and calculating the average value; obtaining the adjustment coefficient at the current moment, and calculating the ratio between it and the average value, and multiplying the initial smoothing coefficient by the ratio to obtain the optimal coefficient.

[0016] As the carburetor runs, the historical data continues to increase, and more adjustment coefficients of historical moments need to be calculated. In the process of calculating the optimal coefficient, this scheme selects some data close to the current moment, which can reduce the amount of calculation and improve the calculation efficiency.

[0017] Optionally, the data selection method of the adjacent data set is: sort the collected multiple historical data according to the collection time to form a time series; determine the number of temperature data in the adjacent data set; select the temperature data close to the current time. The temperature data constitutes a neighboring data set.

[0018] Optionally, the step of predicting the temperature using an exponential smoothing method according to the optimal coefficient includes: monitoring the vaporizer temperature according to the result of the temperature prediction, and alarming when the vaporizer exceeds a preset temperature range.

[0019] In a second aspect, the present application provides a The temperature monitoring system for the vaporizer adopts the following technical solutions: A sort of A temperature monitoring system for a vaporizer, comprising a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the system realizes the above-mentioned Temperature monitoring method for vaporizers.

[0020] One of the above The vaporizer temperature monitoring method generates a computer program and stores it in a memory so as to be loaded and executed by a processor. Thus, a system is made according to the memory and the processor for easy use.

[0021] The present application has the following technical effects: determining the adjacent data set at the current moment, determining the adjustment coefficient according to the fluctuation of the data in the adjacent data set and the change area of ​​the data in the adjacent data set, and adjusting the smoothing coefficient used in the exponential smoothing method according to the adjustment coefficient, so that the temperature change during the use of the vaporizer can be taken into account in the prediction process, thereby improving the responsiveness to real-time temperature changes in the prediction process. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This application is a Method flow chart for temperature monitoring method for vaporizer.

[0023] Figure 2 This application is a Method flow chart of step S1 of the method for monitoring temperature of a vaporizer. DETAILED DESCRIPTION

[0024] The present application embodiment discloses a The temperature monitoring method and system for the vaporizer calculate the optimal coefficient through the temperature fluctuation and temperature change trend in the historical temperature data, and adjust the initial smoothing coefficient in the exponential smoothing method prediction process according to the optimal coefficient, so as to automatically adjust the smoothing coefficient in the exponential smoothing method prediction process and enhance the responsiveness of the exponential smoothing method to the real-time temperature change.

[0025] Reference Figure 1 , The temperature monitoring method for a vaporizer includes steps S1 and S2.

[0026] S1: Calculate the optimal coefficient; Reference Figure 2 , step S1 includes steps S11-S14.

[0027] S11: Obtain temperature data at historical moments, and select a preset number of data to construct a near-current data set at the current moment; During the operation of the vaporizer, the temperature of the vaporizer is collected at a certain frequency, and the vaporizer temperature is predicted after the vaporizer has been running for a certain period of time. In this embodiment, the collection frequency is 60 times per second, and the prediction start time is 20 minutes after the vaporizer has been running. In other embodiments, the frequency of data collection and the prediction start time can be adjusted according to actual conditions; for example, in other embodiments, the collection frequency can be 30 times or 20 times per second. The prediction start time can be 10 minutes or 15 minutes.

[0028] Set the data set capacity, that is, the number of data in the data set, denoted as ; Sort the temperature data in the historical data set according to the time series; extract the historical data close to the current time temperature data, forming a neighboring data set. The temperature setting index value is 1, 2, 3, etc. from the side farthest from the current moment to the side close to the current moment in the adjacent data set. … In this embodiment =60. In other embodiments, the staff can adjust the Make adjustments.

[0029] S12: Calculate the local fluctuation degree and trend influencing factor at the current moment based on the temperature data in the adjacent data set; According to the degree of change of the temperature data in the adjacent data set, the fluctuation of the temperature data in the adjacent data set at the current moment is calculated. According to the trend of the temperature data in the adjacent data set changing over time, the trend impact factor at the current moment is calculated.

[0030] Step S12 includes steps S121 to S123.

[0031] S121: Calculate the weighted average of data in adjacent data sets; The formula for calculating the weighted average is: ; In the formula, Indicates The weighted average of the temperature data sets adjacent to each other; Indicates the number of data in the adjacent data set. is the position index value of the temperature data in the adjacent data set, representing the position of the temperature data in the time series composed of the adjacent data set; Indicates The temperature data in the adjacent data set In the time series composed of temperature data, the position index value corresponding to the temperature data close to the current moment is greater than the index value of the temperature data far from the current moment. Under the influence of the value, the data close to the current moment is assigned more weight to increase the attention to the nearby temperature data.

[0032] S122: Calculation of local fluctuation degree; The local fluctuation degree at the current moment is calculated by comparing different data in the adjacent data set with the weighted average of the data in the adjacent data set.

[0033] In one embodiment, the calculation formula for the local fluctuation degree is: ; In the formula, Indicates The local fluctuation degree of temperature data, Indicates The weighted average of the temperature data sets adjacent to the data set. Indicates the number of data in the adjacent data set. Indicates The temperature data in the adjacent data set Temperature measurements, represents the maximum value function; is an exponential function with a natural constant as its base.

[0034] It should be understood that in this application, the current moment is not absolute, but is divided according to the time series. The temperature data is the current moment, which is the moment of the data to be calculated. The historical moment refers to the The moment before the temperature data.

[0035] Indicates the number of adjacent data sets The difference between each temperature measurement value and the weighted average value of the adjacent data set reflects the fluctuation of the temperature data in the data set through the size of the difference between each data and the weighted average value. The greater the sum of the differences between each temperature data and the weighted average value, the greater the fluctuation of the temperature in the adjacent data set.

[0036] and Make a comparison and normalize the difference to facilitate calculation. The difference between the temperature data in the adjacent data sets and the weighted average is weighted to obtain the local fluctuation degree, so that the calculation of the local fluctuation degree pays more attention to the leap area.

[0037] In another embodiment, the calculation formula for the local fluctuation is: ; Where: Indicates The weighted average of the temperature data sets adjacent to the data set. Indicates the number of data in the adjacent data set. Indicates The temperature data in the adjacent data set Temperature measurements, Indicates The local fluctuation degree of temperature data, represents the maximum value function; is an exponential function with a natural constant as its base.

[0038] This formula uses absolute value calculation to simplify the calculation steps and improve calculation efficiency.

[0039] S123: Calculation of trend impact factors; In historical data, the difference between the measured value of historical data and the predicted value of historical temperature reflects the accuracy of the historical prediction process; the greater the difference between the predicted value and the measured value, the lower the accuracy of the previous prediction. In this case, we should pay more attention to the nearby temperature data. Then, we can calculate the trend influence factor of the temperature data by combining the law of temperature change over time in the nearby data set.

[0040] The trend impact factor formula is: ; Indicates The trend influencing factor of temperature data is Indicates the number of temperature data in the adjacent data set. Indicates The temperature data is adjacent to the Temperature measurements, Indicates The temperature data in the adjacent data set The predicted temperature value, Indicates The temperature data is close to the extreme value of the data set. Indicates The weighted average of the temperature data sets adjacent to the data set.

[0041] The deviation between the actual value and the predicted value during the forecasting process is calculated in part, and the deviation can reflect the accuracy of the forecast value. If the gap is large, it means that the historical forecast results are not accurate enough, and a larger smoothing coefficient is needed to make the forecast model more dependent on the most recent data points; Indicates The range of the data set adjacent to each temperature data set is calculated, and the deviation between the actual value and the predicted value is compared with the range of the data set adjacent to the temperature data set. The calculation result of the deviation is normalized to facilitate subsequent calculations.

[0042] The deviation between the temperature data in the adjacent data set and the weighted average of the adjacent data set and the deviation of the position index value in the sequence of index values ​​are calculated, and the trend of temperature change over time is reflected by the product of the two deviations. In the actual temperature prediction operation process, if the temperature continues to rise or fall, this will cause the value of this formula to become larger. According to the value of this formula, it can be concluded that more attention should be paid to the nearest data points in the prediction process, and the dependence on the predicted value should be reduced to improve the accuracy of temperature prediction.

[0043] S13: The product of the local fluctuation degree and the trend impact factor is used as the adjustment coefficient; In order to predict the temperature more accurately, the temperature at future moments should be predicted in combination with the local fluctuation degree and the trend influencing factor, so that the local fluctuation degree and the trend influencing factor can act together on the adjustment coefficient.

[0044] S14: setting an initial smoothing coefficient, and adjusting the initial smoothing coefficient by adjusting the coefficient to obtain an optimal coefficient; In one embodiment, the optimal coefficient is obtained by the following steps: setting the initial smoothing coefficient in the exponential smoothing forecasting process; calculating the adjustment coefficient at each moment in the historical moment, and then calculating the average of multiple adjustment coefficients. The data of each moment cannot constitute a nearby data set, so it is not calculated. At the same time, as the time continues, new adjustment coefficients of historical moments are generated, and the mean of the adjustment coefficients of historical moments changes accordingly. The average value of the adjustment coefficients of multiple historical moments is calculated, and the initial smoothing coefficient is adjusted by the ratio of the current moment's adjustment coefficient to the average value of the adjustment coefficients of historical moments to obtain the optimal coefficient.

[0045] The calculation formula of the optimal coefficient is: ; In the formula Indicates the adjustment coefficient at the current moment; Represents the average value of adjustment coefficients at multiple historical moments; represents the optimal coefficient; is the initial smoothing coefficient.

[0046] In other embodiments, the temperature data within a certain range from the current moment may be selected to calculate the adjustment coefficient to reduce the computational burden. For example, data within a 20-minute interval from the current moment may be selected for calculation. Alternatively, data within a 10-minute interval from the current moment may be selected for calculation.

[0047] S2: Use exponential smoothing method to predict temperature based on the optimal coefficient; Monitor the vaporizer temperature based on the results of temperature prediction, and give an alarm in time when the vaporizer temperature exceeds the preset temperature range; or when the vaporizer temperature change is inconsistent with the operating status of the vaporizer (the temperature continues to rise after the vaporizer is stopped in history).

[0048] The present application also discloses a The temperature monitoring system for a vaporizer includes a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a method according to the present application is implemented. Temperature monitoring method for vaporizers.

[0049] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, and their configuration and functions are known in the art, so they will not be described in detail here.

[0050] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A A temperature monitoring method for a vaporizer, characterized in that: The method comprises the steps of: calculating an optimal coefficient; Use exponential smoothing method to predict temperature based on the optimal coefficient; The steps of calculating the optimal coefficient include: obtaining temperature data at historical moments, and selecting a preset number of data to construct a current near-data set; calculating the local fluctuation degree and trend influence factor at the current moment according to the temperature data in the near-data set; taking the product of the local fluctuation degree and the trend influence factor as the adjustment coefficient; setting an initial smoothing coefficient, and adjusting the initial smoothing coefficient by the adjustment coefficient to obtain the optimal coefficient; wherein, the calculation formula of the trend influence factor is: ; Where: Indicates The trend influencing factor of temperature data is Indicates the number of temperature data in the adjacent data set. Indicates The temperature data in the adjacent data set Temperature measurements, Indicates The temperature data in the adjacent data set The predicted temperature value, Indicates The temperature data is close to the extreme value of the data set. Indicates The weighted average of the temperature data sets adjacent to the data set.

2. A method according to claim 1 A temperature monitoring method for a vaporizer, characterized in that: The formula for calculating the weighted average of adjacent data sets is: ; In the formula, Indicates The weighted average of the adjacent data sets of temperature data, Indicates the number of data in the adjacent data set. Indicates The temperature data in the adjacent data set Temperature measurements; Indicates the index value of the temperature data in the adjacent dataset.

3. A method according to claim 2 A temperature monitoring method for a vaporizer, characterized in that: The calculation formula for the local fluctuation degree is: ; In the formula, Indicates The weighted average of the temperature data sets adjacent to the data set. Indicates the number of data in the adjacent data set. Indicates The temperature data in the adjacent data set Temperature measurements, Indicates The local fluctuation degree of temperature data, represents the maximum value function; is an exponential function with a natural constant as its base.

4. A method according to claim 2 A temperature monitoring method for a vaporizer, characterized in that: The calculation formula for the local fluctuation degree is: ; In the formula, Indicates The weighted average of the temperature data sets adjacent to the data set. Indicates the number of data in the adjacent data set. Indicates The temperature data in the adjacent data set Temperature measurements, Indicates The local fluctuation degree of temperature data, represents the maximum value function; is an exponential function with a natural constant as its base.

5. A method according to claim 1 A temperature monitoring method for a vaporizer, characterized in that: The step of predicting the temperature using the exponential smoothing method according to the optimal coefficient includes: monitoring the temperature of the vaporizer according to the result of the temperature prediction, and alarming when the temperature change of the vaporizer is contrary to the operating state of the vaporizer.

6. A method according to claim 1 A temperature monitoring method for a vaporizer, characterized in that: The steps of adjusting the initial smoothing coefficient to obtain the optimal coefficient include: obtaining the adjustment coefficient of each moment in the historical moments and calculating the average value; obtaining the adjustment coefficient of the current moment and calculating the ratio between it and the average value, and taking the product of the initial smoothing coefficient and the ratio as the optimal coefficient.

7. A method according to claim 1 A temperature monitoring method for a vaporizer, characterized in that: The steps of adjusting the initial smoothing coefficient to obtain the optimal coefficient include: obtaining the adjustment coefficient of the temperature data with a time interval of less than 20 minutes from the current moment, and calculating the average value; obtaining the adjustment coefficient at the current moment, and calculating the ratio between it and the average value, and taking the product of the initial smoothing coefficient and the ratio as the optimal coefficient.

8. A method according to claim 1 A temperature monitoring method for a vaporizer, characterized in that: The data selection method of the adjacent data set is as follows: the collected multiple historical data are sorted according to the time of collection to form a time series; the number of temperature data in the adjacent data set is determined, recorded as ; Choose the one that is close to the present moment The temperature data constitutes a neighboring data set.

9. A method according to claim 1 A temperature monitoring method for a vaporizer, characterized in that: The step of predicting the temperature using the exponential smoothing method according to the optimal coefficient includes: monitoring the vaporizer temperature according to the result of the temperature prediction, and alarming when the vaporizer exceeds the preset temperature range.

10. A A temperature monitoring system for a vaporizer, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a method according to any one of claims 1 to 9 is implemented. Temperature monitoring method for vaporizers.

Images