Temperature control method and system for natural convection constant temperature test box

Abstract

The invention relates to the field of data processing, in particular to a temperature control method and system for a natural convection constant-temperature test box, and the method comprises the steps: collecting the internal temperature of the test box, and calculating the abnormal degree of each temperature; calculating the difference of all the temperatures, taking the difference as an error change rate, and taking the product of the abnormal degrees of two adjacent temperatures corresponding to the error change rate as the abnormal degree of the error change rate; adjusting the error change rate based on the abnormal degree of the error change rate to obtain an adjusted error change rate; the current temperature in the test box is obtained, the adjusted error change rate corresponding to the current temperature is calculated, the current temperature and the adjusted error change rate corresponding to the current temperature are input into a PID control algorithm, and temperature control is conducted on the test box according to the output result of the PID control algorithm. According to the invention, through calculation of the temperature abnormal degree and adjustment of the error change rate, control of the internal temperature of the test box is more accurate.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and more specifically, to a temperature control method and system for a natural convection constant temperature test box. Background Art

[0002] The natural convection constant temperature test chamber is a test equipment used to simulate the actual indoor ambient temperature. Through the natural convection test equipment and software, a constant temperature environment (natural convection) is generated that does not require forced air circulation through fans, and relevant temperature detection and test integration are performed on the products to be tested. The natural convection constant temperature test chamber is used in the electronics, electrical appliances, automobiles, aerospace and other industries to test the performance and reliability of various electronic products or confined spaces under actual ambient temperatures.

[0003] Prior art, such as a patent application document with publication number CN109375684A, discloses a PID control method, which includes: collecting the temperature and humidity in the system; controlling the start-up time of the PTC heater, circulating water pump, and circulating fan in the system according to the set temperature and set humidity, as well as the collected current temperature and current humidity; calculating the difference between the set temperature and the current temperature, and the difference between the set humidity and the current humidity, and using these differences and an incremental PID algorithm to adjust the output power of the PTC heater, the circulating water pump, and the circulating fan.

[0004] Although the above PID control method has achieved effective regulation of system temperature and humidity to a certain extent, it still has some limitations. In particular, in the process of temperature data collection, external factors such as high-frequency noise in the measurement data and system dynamic changes will affect the calculation results of the PID controller, resulting in reduced temperature control accuracy. Summary of the invention

[0005] In order to solve the above technical problem that the accuracy of temperature control is reduced due to interference from external factors, the present invention provides solutions in the following aspects.

[0006] In a first aspect, a temperature control method of a natural convection constant temperature test chamber comprises: Collect the temperature inside the test chamber and calculate the abnormal degree of each temperature; Calculate the difference of all temperatures, take the difference as the error change rate, and take the product of the abnormality degrees of two adjacent temperatures corresponding to the error change rate as the abnormality degree of the error change rate; The error change rate is adjusted based on the abnormal degree of the error change rate to obtain an adjusted error change rate; The current temperature inside the test chamber is obtained, the adjusted error change rate corresponding to the current temperature is calculated, the current temperature and the adjusted error change rate corresponding to the current temperature are input into the PID control algorithm, and the temperature of the test chamber is controlled according to the output result of the PID control algorithm.

[0007] By collecting the internal temperature of the test box and calculating the abnormal degree of each temperature, the present invention can accurately identify abnormal points in the temperature data, thereby performing more detailed monitoring of the temperature change and avoiding test errors caused by the collected temperature abnormalities; the product of the abnormal degrees of two adjacent temperatures corresponding to the error change rate is used as the abnormal degree of the error change rate, and the error change rate is adjusted based on this, so that the error change rate is closer to the actual temperature change situation, effectively reducing the problem of false high differential terms caused by external factors in the PID control method, and improving the accuracy and stability of temperature control.

[0008] According to the change of the temperature inside the test chamber, the error change rate is automatically adjusted to adapt to different temperature change rates. Whether it is a rapid heating, rapid cooling or slow temperature change process, accurate control can be achieved through the synergy of the adjusted error change rate and the PID control algorithm, expanding the scope of application of temperature control.

[0009] Preferably, the process of obtaining the abnormality degree includes: Calculate the similarity between any two temperatures; Clustering algorithms are used to cluster all collected temperatures to obtain multiple clusters; The size of the area with no data points around each temperature in the cluster and the average distribution range of the temperature in the cluster are calculated, and the ratio of the size of the area with no data points around the temperature to the average distribution range is used as the abnormality degree of the corresponding temperature; wherein the size of the area with no data points around the temperature represents the area size of a circle with any temperature in the cluster as the center point and a radius as the minimum similarity between the center point and other temperatures, and no other temperature exists in the area.

[0010] By calculating the similarity between temperatures, clustering, and calculating the degree of abnormality, the abnormal points in the temperature data can be effectively identified. These abnormal points may be caused by noise caused by data acquisition and transmission, so these abnormal data can be marked to reduce their impact on the PID algorithm.

[0011] Preferably, the process of obtaining the abnormality degree further includes: Use the exponential smoothing method to predict all temperatures and obtain the predicted value corresponding to each temperature; Calculate the standard deviation of all temperatures; The ratio of the difference between the actual value and the predicted value of each temperature to the standard deviation is used as the predicted abnormality degree of the corresponding temperature; The predicted abnormality degree is used as a correction factor, and the product of the correction factor and the ratio of the size of the area with no data points around the temperature to the average distribution range is used as the final abnormality degree.

[0012] The introduction of the predicted abnormality degree is equivalent to providing a dynamically adjusted threshold for anomaly detection. When the temperature data changes relatively smoothly and the noise is small, the predicted abnormality degree is small, and the threshold of anomaly detection is relatively low, which can more sensitively detect small abnormal changes; when the data changes drastically and the noise is large, the predicted abnormality degree is large, and the threshold of anomaly detection is correspondingly increased to avoid oversensitivity and misjudgment caused by noise, so that the anomaly detection method can automatically adjust the detection strategy according to the dynamic characteristics of the data.

[0013] Preferably, the adjusted error change rate satisfies the relationship: ; In the formula, Indicates The weighted error change rate, Indicates The abnormal degree of the error change rate, Indicates The value of the error change rate, Indicates The error rate of change to The sum of the inverses of the error change rates, Indicates The error rate of change The weighted sum of the error change rates.

[0014] By weighting the error change rate, the error change rate with a smaller degree of abnormality (i.e., more normal data) has a larger weight in the calculation, while the error change rate with a larger degree of abnormality has a smaller weight. This can reduce the impact of abnormal data on the overall error change rate, so that the calculation results can better reflect the error change trend under normal conditions, thereby improving control accuracy.

[0015] In temperature control systems, noise is often introduced during data collection and transmission, resulting in fluctuations in the error rate of change. By introducing the inverse of the degree of abnormality as a weight, the impact of noise data on the error rate of change calculation can be effectively reduced, making the control system more stable in the face of noise interference, reducing false operations and unnecessary adjustments, and enhancing the robustness of the system.

[0016] Preferably, the output result of the PID control algorithm is: ; In the formula, is the control signal, is the proportionality coefficient, is the integration coefficient, is the differential coefficient, is the error between the current temperature and the set temperature of the test chamber. is the weighted error change rate corresponding to the current temperature.

[0017] In traditional PID control algorithms, the differential term mainly reflects the changing trend of the error, and is used to predict future errors and make adjustments in advance to reduce system overshoot and speed up system response. After the introduction of the weighted error change rate, the differential term can more accurately reflect the dynamic characteristics of the error change, because the weighted error change rate takes into account the weighted sum of the first L error change rates, and can better capture the trend and speed of the error change. This enables the control system to respond more quickly to temperature changes and adjust the control signal in advance, thereby speeding up the system's response and reducing the time required for the temperature to reach the set value.

[0018] Preferably, controlling the temperature of the test box according to the output result of the PID control algorithm includes: If the output result is positive, increase the heating power of the test chamber; otherwise, reduce the heating power of the test chamber.

[0019] Preferably, the clustering algorithm is a K-Means clustering algorithm.

[0020] In a second aspect, a temperature control system of a natural convection constant temperature test chamber includes: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the temperature control method of the natural convection constant temperature test chamber is implemented.

[0021] The beneficial effects of the present invention are: Since the presence of noise can cause fluctuations in temperature data, the present invention accurately identifies abnormal points by calculating the abnormal degree of the temperature inside the test chamber, thereby improving the temperature monitoring accuracy and reducing test errors. At the same time, it automatically adjusts according to the abnormal degree of the error change rate, combined with the PID control algorithm, to achieve more accurate temperature control, adapt to different temperature change rates, and expand the scope of application of temperature control. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein: Figure 1It is a method flow chart of steps S1 to S4 in a temperature control method of a natural convection constant temperature test box according to an embodiment of the present invention. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0024] The application scenario of the present invention is: using a PID control algorithm to regulate a natural convection constant temperature test box.

[0025] The embodiment of the present invention discloses a temperature control method of a natural convection constant temperature test box, referring to Figure 1 , including steps S1 to S4, which are specifically as follows: S1: Collect the temperature inside the test chamber and calculate the abnormality of each temperature.

[0026] PID algorithm is a commonly used control algorithm, which adjusts the system input by performing proportional, integral and differential operations on the error between the system output and the set value so that the system output reaches the expected value. However, when there are abnormal data in the system, these abnormal data will cause deviations in the error calculation, thus affecting the control effect of the PID algorithm.

[0027] Therefore, in order to improve the accuracy and stability of the PID algorithm, it is necessary to monitor the temperature inside the test chamber in real time and calculate the abnormal degree of each temperature. By identifying and processing abnormal data, its impact on the PID algorithm can be reduced, thereby improving the performance of the control system.

[0028] Specifically, a temperature sensor is set inside the test box, and then the temperature sensor is used to collect the temperature inside the test box in real time, that is, the temperature data is collected once a minute, at least 60 times, and then the collected temperature data is preprocessed, including data cleaning and data smoothing.

[0029] Furthermore, the abnormal degree of temperature data is calculated. The specific steps are as follows: First, since temperature data usually has time dependence, that is, the temperature value is not only affected by time, but the temperature values ​​at adjacent time points usually have a high correlation, therefore, considering the similarity calculation of time position can more accurately reflect the similarity between two temperature data.

[0030] Specifically, the similarity between two temperature data satisfies the relationship:

[0031] In the formula, For the Second and The similarity between the temperatures collected, For the The temperature value collected is For the The temperature value collected is For the The result after normalization of the index values ​​collected. For the The result after normalization of the index values ​​collected. is the total number of collections.

[0032] in, , This directly reflects the temperature difference between the two acquisitions. and It reflects the position difference of the two acquisitions in the time series. By combining these two factors, not only the difference in the temperature values ​​themselves but also the position difference of these temperature values ​​in the time series is considered, which helps to more comprehensively evaluate the similarity of two data points.

[0033] Then, the adaptive K-Means clustering algorithm is used to cluster the temperature data and generate multiple clusters. This is because the traditional K-Means algorithm requires the number of clusters to be specified in advance. However, in practical applications, it is not known in advance how many clusters should be divided into. The adaptive K-Means clustering can automatically determine the optimal number of clusters through iterative calculation and adjustment, thereby avoiding the deviation that may be caused by the artificially set number of clusters.

[0034] Each of the above clusters has a center, which represents the average temperature and average collection index of all temperature data in the cluster.

[0035] For If a temperature data is far away from the center of the cluster, it may mean that the temperature data is very different from other temperature data in the cluster.

[0036] To quantify this difference, we first find The minimum and maximum values ​​of the temperature data within the cluster, the minimum and maximum values ​​of the collection index, for the The first temperature data, and calculate the minimum similarity between the temperature data and other temperature data in the cluster to which it belongs.

[0037] Further calculation of The first The abnormal degree of temperature data is the relationship that satisfies:

[0038] In the formula, For the The first The abnormality of the temperature data is is the circumference of a circle, For the The first The minimum similarity between a temperature data and other temperature data, For the The maximum temperature value in the clusters, For the The minimum temperature value in the clusters, For the The maximum value of the collection index in the clusters, For the The minimum value of the collection index in the clusters, For the The total number of temperature data in the clusters.

[0039] Among them, the molecule Indicates The first The size of the area without data points around the temperature data, that is, The first The temperature data is the center and the radius is The size of the circle, where no other data exists; if there is a large area with no data points around a temperature data, it means that the data is very different from other data, and the data is more likely to be abnormal. Indicates The average distribution range of temperature data in the clusters.

[0040] The larger the The first The individual temperature data are very discrete relative to the other temperature data in their cluster.

[0041] According to the above The calculation method of is similar to that of the abnormal degree of all other temperatures.

[0042] In addition, it is important to consider that temperature data is usually time series data, that is, the data changes over time. Similarity calculations and cluster analysis based solely on the static features of data points (such as temperature values ​​and collection indexes) may not fully capture the dynamic characteristics of data changes over time. By predicting the degree of anomaly calculation, we can use time series forecasting techniques such as exponential smoothing to predict temperature data and evaluate the difference between actual and predicted values. This difference reflects the degree of change in the data over time and helps identify anomalies that deviate significantly from the expected trend.

[0043] Specifically, all collected temperature data are predicted using exponential smoothing to obtain the predicted value corresponding to each temperature data, and then the standard deviation of all temperature data is calculated.

[0044] For each temperature data, the difference between the actual value and the predicted value is calculated, and this difference is divided by the standard deviation to obtain the predicted abnormality degree of the temperature data. This predicted abnormality degree reflects the degree of deviation between the actual temperature and the predicted temperature. The greater the deviation, the more likely the temperature data is abnormal.

[0045] Furthermore, the calculated predicted abnormality degree is used as a correction factor and multiplied by the ratio of the size of the data-free area around the temperature calculated above to the average distribution range to obtain the final abnormality degree, that is, the relationship is satisfied:

[0046] In the formula, For the The final abnormal degree of temperature, For the The degree of abnormality of a temperature without the correction factor added, For the The correction factor for temperature.

[0047] This final degree of anomaly comprehensively considers the predicted anomaly of the temperature data in time series and the degree of discreteness in space, thereby being able to more comprehensively reflect the abnormality of the temperature data.

[0048] S2: Calculate the difference of all temperatures, take the difference as the error change rate, and take the product of the abnormality levels of two adjacent temperatures corresponding to the error change rate as the abnormality level of the error change rate.

[0049] First, calculate the difference of all temperatures, which can be expressed as: ,in, It is The difference value (i.e. error change rate), For the Temperature value, For the A temperature value.

[0050] It can be used as the error change rate to indicate the change of temperature data at adjacent moments. The error change rate is the basis of the differential term in the PID control algorithm and is used to reflect the change trend of the temperature error.

[0051] In the above S1, the abnormality of each temperature data has been calculated (calculated based on the similarity of the temperature data, clustering results, prediction error and other information, used to measure whether each temperature data is affected by noise or other abnormal factors). In order to reduce the impact of abnormal data on the error change rate, the above The product of the abnormal degree of the two corresponding temperatures is given by The abnormal degree of , that is, the relationship is satisfied:

[0052] In the formula, For the The abnormal degree of the error change rate, For the The abnormal degree of temperature, For the The degree of temperature abnormality.

[0053] S3: adjusting the error change rate based on the abnormality degree of the error change rate to obtain an adjusted error change rate.

[0054] After the above S2 calculates the abnormal degree of each error change rate, the influence of the abnormal value on the error change rate calculation is further reduced by the weighted summation method, so that the final weighted error change rate is more stable and reliable.

[0055] Specifically, take the The weighted average of the L error change rates before the error change rate is obtained The weighted error rate of change.

[0056] The error rate of change reflects the trend of data over time. By considering multiple previous error rates of change, we can better capture and understand the changes in this trend. For example, if the recent error rate of change shows a gradually increasing trend, it may mean that the system is undergoing some kind of change or adjustment. Through weighted processing, this trend can be quantified more accurately and more informed decisions can be made accordingly.

[0057] For example, for For each error change rate, intercept the 20 previous error change rates and calculate the sum of the inverses of the abnormality of these error change rates; for the The 20 error change rates before the first error change rate are taken as their weights, and the weighted sum of these 20 error change rates is calculated; the weighted sum is divided by the sum of the inverses of the abnormality levels to obtain the The weighted error change rate. That is, the relationship is satisfied:

[0058] In the formula, Indicates The weighted error change rate, Indicates The abnormal degree of the error change rate, Indicates The value of the error change rate, Indicates The error rate of change to The sum of the inverses of the error change rates, Indicates The error rate of change The weighted sum of the error change rates.

[0059] By the above method, we can obtain A weighted error change rate, which takes into account the previous The abnormal degree of the error change rate is reduced, thereby reducing the impact of abnormal data on the error change rate.

[0060] S4: Obtain the current temperature inside the test chamber, calculate the adjusted error change rate corresponding to the current temperature, input the current temperature and the adjusted error change rate corresponding to the current temperature into the PID control algorithm, and control the temperature of the test chamber according to the output result of the PID control algorithm.

[0061] The current temperature data is collected by the temperature sensor, and the collected current temperature data is compared with the set temperature of the test chamber to obtain the error value; and the corresponding weighted error change rate of the current temperature is recalculated according to the above-mentioned method of calculating the weighted error change rate.

[0062] The current temperature and the weighted error change rate of the current temperature are passed as input parameters to the PID control algorithm. The PID control algorithm calculates the control signal based on the input temperature error and the weighted error change rate, and adjusts the heating power of the test chamber based on the calculated control signal. The larger the output value of the control signal, the more heating power needs to be increased; the smaller the control output value, the more heating power needs to be reduced. For example, if the control output is positive, the heating power is increased; if the control output is negative, the heating power is reduced.

[0063] Among them, the control signal satisfies the relationship:

[0064] In the formula, is the control signal, is the proportionality coefficient, is the integration coefficient, is the differential coefficient, is the error between the current temperature and the set temperature of the test chamber. is the weighted error change rate corresponding to the current temperature.

[0065] Among them, the weighted temperature error change rate is used Replaces the traditional error rate of change PID control is performed. By using the weighted temperature error change rate, the impact of temperature data with a large degree of abnormality on the PID control results can be reduced. This is because the weighted processing ensures that data with a large degree of abnormality (i.e. data that is more affected by noise) will be given a smaller weight, while data with a low degree of abnormality (i.e. more reliable data) will be given a larger weight. In this way, the weighted error change rate can better reflect the true change trend of the temperature data and reduce the impact of abnormal data on the control results.

[0066] In summary, in the process of using the PID algorithm to control the temperature of a natural convection constant temperature test chamber, when a situation occurs that is obviously inconsistent with the conventional error change rate, causing the differential term in the PID algorithm to be artificially high, by introducing historical error data and the error change rate, the degree of abnormality of each error data is calculated, and the weights of the most recent multiple errors are calculated according to the degree of abnormality for weighted operations to obtain a more accurate discrete temperature data differential term, thereby obtaining a more accurate PID calculation result.

[0067] An embodiment of the present invention further discloses a temperature control system for a natural convection constant temperature test box, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a temperature control method for the natural convection constant temperature test box according to the present invention is implemented.

[0068] The system also includes other components familiar to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and will not be described in detail here.

[0069] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, apparatus or device. For example, a computer-readable storage medium may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random-Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high-bandwidth memory HBM (High-Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module or both. Any such computer storage medium may be part of a device or accessible or connectable to a device.

[0070] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0071] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A temperature control method for a natural convection constant temperature test chamber, characterized in that: include: Collect the temperature inside the test chamber and calculate the abnormal degree of each temperature; Calculate the difference of all temperatures, take the difference as the error change rate, and take the product of the abnormality degrees of two adjacent temperatures corresponding to the error change rate as the abnormality degree of the error change rate; The error change rate is adjusted based on the abnormal degree of the error change rate to obtain an adjusted error change rate; The current temperature inside the test chamber is obtained, the adjusted error change rate corresponding to the current temperature is calculated, the current temperature and the adjusted error change rate corresponding to the current temperature are input into the PID control algorithm, and the temperature of the test chamber is controlled according to the output result of the PID control algorithm.

2. The temperature control method of a natural convection constant temperature test chamber according to claim 1, characterized in that: The process of obtaining the abnormality degree includes: Calculate the similarity between any two temperatures; Clustering algorithms are used to cluster all collected temperatures to obtain multiple clusters; The size of the area with no data points around each temperature in the cluster and the average distribution range of the temperature in the cluster are calculated, and the ratio of the size of the area with no data points around the temperature to the average distribution range is used as the abnormality degree of the corresponding temperature; wherein the size of the area with no data points around the temperature represents the area size of a circle with any temperature in the cluster as the center point and a radius as the minimum similarity between the center point and other temperatures, and no other temperature exists in the area.

3. The temperature control method of a natural convection constant temperature test chamber according to claim 2, characterized in that: The process of obtaining the abnormality degree further includes: Use exponential smoothing method to predict all temperatures and get the predicted value corresponding to each temperature; Calculate the standard deviation of all temperatures; The ratio of the difference between the actual value and the predicted value of each temperature to the standard deviation is used as the predicted abnormality degree of the corresponding temperature; The predicted abnormality degree is used as a correction factor, and the product of the correction factor and the ratio of the size of the area with no data points around the temperature to the average distribution range is used as the final abnormality degree.

4. The temperature control method of a natural convection constant temperature test chamber according to claim 3, characterized in that: The error change rate after adjustment satisfies the relationship: ; In the formula, Indicates The weighted error change rate, Indicates The abnormal degree of the error change rate, Indicates The value of the error change rate, Indicates The error rate of change to The sum of the inverses of the error change rates, Indicates The error rate of change The weighted sum of the error change rates.

5. The temperature control method of a natural convection constant temperature test box according to claim 4, characterized in that: The output of the PID control algorithm is: ; In the formula, is the control signal, is the proportionality coefficient, is the integration coefficient, is the differential coefficient, is the error between the current temperature and the set temperature of the test chamber. is the weighted error change rate corresponding to the current temperature.

6. A temperature control method for a natural convection constant temperature test chamber according to claim 5, characterized in that: Temperature control of the test chamber based on the output of the PID control algorithm includes: If the output result is positive, increase the heating power of the test chamber; otherwise, reduce the heating power of the test chamber.

7. The temperature control method of a natural convection constant temperature test chamber according to claim 2, characterized in that: The clustering algorithm is K-Means clustering algorithm.

8. A temperature control system for a natural convection constant temperature test chamber, 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, the temperature control method of the natural convection constant temperature test box according to any one of claims 1 to 7 is implemented.

Images