Temperature control method of electric control cabinet based on artificial intelligence

Through the temperature control method of the electric control cabinet based on artificial intelligence, the importance of the temperature sensor is determined and screened, the problem of unstable temperature control of the electric control cabinet is solved, the accuracy and reliability of temperature regulation are improved, and the electrical control cabinet is guaranteed security.

CN118585011BActive Publication Date: 2024-11-29WUXI KANGBEI ELECTRONICS EQUIP
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202411035609.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-31
Publication Date
2024-11-29
Estimated Expiration
2044-07-31

AI Technical Summary

Technical Problem

During the operation of the electrical control cabinet, the temperature control is unstable due to the influence of environmental conditions. In the cluster analysis, the existing technology has poor clustering effect due to too many parameters, which leads to temperature regulation errors.

Method used

Using the temperature regulation method of the electric control cabinet based on artificial intelligence, determine the steps of the importance of the temperature sensor, obtain the importance of the internal and external temperature sensors, and obtain important characteristic parameters through importance screening, and cluster to improve the accuracy of temperature regulation. Sex and reliability.

Benefits of technology

By analyzing the correlation and abnormality of temperature parameters, redundant parameters are reduced, the accuracy and reliability of temperature adjustment are improved, and the safety of the electrical control cabinet is ensured.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118585011B_ABST
    Figure CN118585011B_ABST
Patent Text Reader

Abstract

The present application relates to the field of temperature control, and specifically to a temperature control method for an electric control cabinet based on artificial intelligence. The method includes: collecting temperatures at different locations at different times, obtaining the importance of each temperature sensor inside the electric control cabinet through the correlation between the temperature sensors inside the electric control cabinet, the correlation with the control temperature, the abnormality of the temperature sensor, and the temperature distribution characteristics and significant characteristics of the temperature sensor; determining the importance of obtaining external temperature sensors through the temperature difference and correlation between the outside and inside of the electric control cabinet, selecting sensors with high importance based on the importance of the inside and outside, and determining the control temperature based on this, and controlling the equipment through the control temperature. The present application controls the temperature of the temperature control equipment to the final control temperature, thereby ensuring the safety of the electric control cabinet.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of temperature control, and specifically to a temperature control method for an electric control cabinet based on artificial intelligence. Background Art

[0002] During the operation of the electric control cabinet, temperature control is required to ensure that the internal equipment does not overheat. When the temperature is too high, the equipment needs to be cooled down, but the influence of some environmental conditions (such as the surrounding environment, airflow, etc.) causes unstable temperature control, which increases the difficulty of temperature control of the electric control cabinet.

[0003] The temperature of the electric control cabinet is regulated based on artificial intelligence. When clustering analysis is performed based on the collected parameters according to historical data, the clustering effect may be poor due to too many unnecessary parameters in the clustering parameters, which may lead to errors in the temperature adjustment of the corresponding electric control cabinet based on the clustering results, resulting in abnormal temperature of the electric control cabinet. Therefore, the temperature of the electric control cabinet needs to be adjusted. Summary of the invention

[0004] In order to solve the technical problem of temperature control errors, this application provides an electric control cabinet temperature control method based on artificial intelligence. The technical solutions adopted are as follows:

[0005] This application proposes an electric control cabinet temperature control method based on artificial intelligence, which includes the following steps:

[0006] Temperature collection steps: collect the temperature of temperature sensors at different locations at each moment;

[0007] Steps to determine the importance of the temperature sensor inside the electric control cabinet:

[0008] Obtain the controlled temperature at each historical moment, and obtain the first importance of the temperature sensor according to the correlation between each temperature sensor and all the temperatures of other temperature sensors, and the correlation between each temperature sensor and the controlled temperature at each historical moment;

[0009] Detecting abnormal points of all temperatures of the temperature sensor, and determining the second importance of the temperature sensor according to the time difference of all abnormal points of the temperature sensor and the temperature difference of the temperature sensor and other temperature sensors at the same time;

[0010] Determining the third importance of the temperature sensor based on the significance of all the temperatures of each temperature sensor and the temperature difference between the temperature sensor and the rest of the surrounding temperature sensors;

[0011] The first importance, second importance, and third importance of the temperature sensor are forwardly integrated to obtain the importance of the temperature sensor inside the electric control cabinet;

[0012] Steps to determine the importance of the external temperature sensor of the electric control cabinet:

[0013] The importance of the temperature sensor outside the electric control cabinet is obtained based on the difference between all temperatures collected by the temperature sensors outside the electric control cabinet and inside the electric control cabinet and the correlation of all controlled temperatures;

[0014] Temperature sensor screening steps:

[0015] Select the temperature sensor with high importance according to the difference in correlation between the temperature values ​​of the internal temperature sensor and the external temperature sensor of the electric control cabinet and the control temperature;

[0016] Temperature control steps: determine the control temperature through a high-importance temperature sensor, and control the temperature of the device to a value corresponding to the control temperature.

[0017] In the above scheme, by analyzing the temperature parameters obtained at different positions inside the electric control cabinet, the influence of the internal parameters on the adjustment results is obtained, so that the parameters with similar effects are suppressed, which greatly reduces the number of redundant parameters; at the same time, based on the historical temperature sequence of the same position, the temperature change stability of the parameters at this position is obtained, which can avoid the situation where the temperature control is inaccurate due to ignoring the parameter data that may suddenly change, and increase the reliability and safety of the control; by analyzing the temperature changes of the temperature parameters and the temperature parameters around them, the uniqueness of the parameter distribution of the temperature parameters at different positions is obtained, which can highlight the importance of high temperature information and increase the accuracy of temperature control; based on the temperature difference inside and outside the electric control cabinet, the importance of the parameters obtained at different external positions is obtained, and the temperature control level can be determined more comprehensively. Finally, based on the importance of temperature parameters, important characteristic parameters are screened and obtained, and clustering is performed based on the characteristic parameters, which greatly increases the accuracy and reliability of obtaining the temperature level of the electric control cabinet, and greatly guarantees the safety of the electric control cabinet.

[0018] In one embodiment, the method for obtaining the first importance of the temperature sensor according to the correlation between each temperature sensor and all the temperatures of other temperature sensors and the correlation between each temperature sensor and the control temperature at the historical moment is:

[0019] At the historical moment, for each temperature sensor, all the temperature values ​​collected by it are sequenced in time order and recorded as a temperature sequence, and the correlation value between each temperature sequence and the other temperature sequences is calculated and recorded as the first correlation value; all the control temperatures corresponding to the historical data are sequenced in time order and recorded as a control sequence, and the correlation value between each temperature sequence and the control sequence is calculated and recorded as the second correlation value;

[0020] Obtaining the importance factor of the temperature sensor according to the correlation values ​​between the temperature sequences and the correlation values ​​between the temperature sequences and the control sequences, wherein the importance factor of the temperature sensor is the first importance;

[0021] Any temperature sequence is recorded as a target sequence, and its corresponding temperature sensor is the target sensor. The importance factor of each target sensor is positively correlated with the first correlation value between the target sequence and the remaining temperature sequences and the second correlation value corresponding to the target sequence; the importance factor of each target sensor is negatively correlated with the second correlation value corresponding to the remaining temperature sequences and the difference between the second correlation values ​​of the target sequence and the remaining temperature sequences.

[0022] In one embodiment, the method for obtaining the importance factor of the temperature sensor according to the correlation values ​​between the temperature sequences and the correlation values ​​between the temperature sequences and the control sequences is:

[0023] The product of the first correlation value between the target sequence and each temperature sequence and the second correlation value corresponding to the target sequence is taken as the temperature importance; the product of the absolute value of the difference between the target sequence and the second correlation value corresponding to each temperature sequence and the second correlation value corresponding to each temperature sequence is recorded as the temperature influence value; after accumulating the ratio of the temperature importance and the temperature influence value of the target sequence and each temperature sequence, the normalized result is taken as the importance factor of the temperature sensor corresponding to the target sequence.

[0024] In one embodiment, the method for determining the second importance of the temperature sensor according to the time difference of all abnormal points of the temperature sensor and the temperature difference between the temperature sensor and other temperature sensors at the same time is:

[0025] For each target sequence, calculate the time interval between each abnormal point and its adjacent abnormal points, take the mean of the time intervals as the time difference of each abnormal point, and construct a sequence of the time differences of all abnormal points as a time regularity sequence;

[0026] Calculating the variance of all time differences in the time regularity sequence, and using the variance as the time regularity value of the time regularity sequence;

[0027] For the target sequence, the abnormal degree of temperature difference change is obtained according to the temperature difference between the target sequence and each other temperature sequence at each moment;

[0028] The abnormal importance of the temperature sensor is determined according to the time regularity value of the temperature sensor and the abnormal degree of the temperature difference change, and the abnormal importance of the temperature sensor is the second importance;

[0029] The importance of anomaly is positively correlated with the time regularity value of the temperature sensor and the abnormality of the temperature difference change of the temperature sensor.

[0030] In one embodiment, the method for obtaining the abnormal degree of temperature difference change according to the temperature difference at each moment between the target sequence and each of the remaining temperature sequences is:

[0031] For the target sequence, the absolute value of the temperature difference between the target sequence and each of the other temperature sequences at each moment is calculated to form a temperature difference sequence between the target sequence and each temperature sequence. The variance of the absolute value of all temperature differences in the temperature difference sequence is calculated, and the inverse of the variance is recorded as the temperature difference change abnormality factor. The average value of the temperature difference change abnormality factor between the target sequence and all temperature sequences is taken as the temperature difference change abnormality.

[0032] In one embodiment, the method for determining the abnormal importance of the temperature sensor according to the time regularity value and the temperature difference value abnormality of the temperature sensor is:

[0033] The abnormal importance of the temperature sensor is obtained by multiplying the time regularity value of the temperature sensor and the abnormality degree of the temperature difference change of the temperature sensor.

[0034] In one embodiment, the method for determining the third importance of the temperature sensor according to the significance of all temperatures of each temperature sensor and the temperature difference between the temperature sensor and the rest of the surrounding temperature sensors is:

[0035] All temperature values ​​of each temperature sensor are linearly normalized as the significance corresponding to the temperature value, and the average of all significances of the temperature sensor is taken as the temperature significance value of the temperature sensor;

[0036] Any temperature sensor is recorded as the central sensor, and a two-dimensional coordinate system is constructed with it as the center. All temperature sensors are divided into four quadrants. The temperature sensors closest to the central sensor are selected from the four quadrants respectively, and these temperature sensors are recorded as neighborhood sensors.

[0037] Determine the degree of temperature change of the central sensor based on the difference between all temperature values ​​of the central sensor and neighboring sensors;

[0038] The degree of temperature variation is positively correlated with the difference between all temperature values ​​of the central sensor and each neighboring sensor;

[0039] Obtaining the temperature distribution importance of the temperature sensor according to the temperature variation degree and the temperature significance value of each temperature sensor, wherein the temperature distribution importance of the temperature sensor is the third importance of the temperature sensor;

[0040] The importance of temperature distribution of temperature sensors is positively correlated with the degree of temperature change and the temperature significance value.

[0041] In one embodiment, the method for forward fusing the importance of the temperature sensor based on the importance of correlation, the importance of anomaly, the distribution characteristics and the importance of significant characteristics to obtain the importance of the temperature sensor inside the electric control cabinet is:

[0042] Obtain the internal importance of each temperature sensor according to the importance factor, abnormal importance, and temperature distribution importance of the temperature sensor;

[0043] The internal importance of the temperature sensor is positively correlated with the importance factor, anomaly importance, and temperature distribution importance of the temperature sensor, respectively.

[0044] In one embodiment, the method for obtaining the importance of the temperature sensor outside the electric control cabinet according to the difference between all temperatures collected by the temperature sensors outside the electric control cabinet and inside the electric control cabinet and the correlation of all controlled temperatures is:

[0045] Inside the electric control cabinet, all temperatures at each moment are divided into two parts, and all temperature values ​​in the part with the higher mean temperature value constitute a high temperature sequence. The temperature of any temperature sensor outside the electric control cabinet at the same moment is subtracted from each temperature in the high temperature sequence, and the mean of all differences at each moment is taken as the internal and external temperature difference of the electric control cabinet at each moment. The sequence formed by the internal and external temperature differences of each temperature sensor outside the electric control cabinet at all moments is recorded as the internal and external temperature difference sequence;

[0046] The correlation value between the calculated temperature internal and external difference sequence and the control sequence is recorded as the third correlation value, and the third correlation value is used as the external importance of the external temperature sensor of the electric control cabinet.

[0047] In one embodiment, the method for selecting a temperature sensor with high importance according to the difference in correlation between the temperature values ​​of the temperature sensor inside the electric control cabinet and the temperature sensor outside the electric control cabinet and the control temperature is:

[0048] For each temperature sensor outside the electric control cabinet, obtain the temperature sequence corresponding to the temperature sensor, calculate the correlation value between the temperature sequence and the control sequence and record it as external correlation;

[0049] For the interior of the electric control cabinet, the correlation value between the temperature sequence of each temperature sensor in the electric control cabinet and the control sequence is divided into two categories using the threshold segmentation method, and the mean of the correlation value of each category is calculated. The mean of the correlation value corresponding to the category with the larger mean of the correlation value is recorded as the internal correlation;

[0050] The ratio of external correlation to internal correlation was used as the suppression factor;

[0051] The product of the suppression factor and the external importance of the external temperature sensor of the electric control cabinet is normalized as the importance screening value;

[0052] A screening threshold is preset, and temperature sensors whose importance screening values ​​are greater than the screening threshold are regarded as important sensors. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0054] Figure 1 A flow chart of an artificial intelligence-based temperature control method for an electric control cabinet provided in one embodiment of the present application;

[0055] Figure 2 The figure is a flowchart of the steps of the temperature control method of the electric control cabinet based on artificial intelligence. DETAILED DESCRIPTION

[0056] In order to further explain the technical means and effects adopted by this application to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of the temperature control method of the electric control cabinet based on artificial intelligence proposed in this application, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0057] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0058] Embodiment of the temperature control method of electric control cabinet based on artificial intelligence:

[0059] The specific scheme of the temperature control method of the electric control cabinet based on artificial intelligence provided by the present application is described in detail below with reference to the accompanying drawings.

[0060] See also Figure 1 , which shows a flow chart of an electric control cabinet temperature control method based on artificial intelligence provided by an embodiment of the present application, the method comprising the following steps:

[0061] S001, collecting the temperature of temperature sensors at different positions at each moment.

[0062] Since the temperature of components at different positions inside the electric control cabinet is different, multiple temperature sensors are set inside and outside the electric control cabinet, where each temperature sensor collects temperature once after a preset time. A temperature control device is set outside the electric control cabinet, and the relative position of the temperature control device and the electric control cabinet remains unchanged, that is, the wind direction of the temperature control device when it is outgoing is fixed. For each temperature sensor, find the part with a high correlation with the temperature control device, and control the temperature of the temperature control device through the correlation between the two, thereby affecting the temperature of the electric control cabinet.

[0063] Preferably, in one embodiment of the present application, the temperature sensor collects data every 10 minutes, wherein the temperature sensor is installed in the middle of the four upper, lower, left and right surfaces of the outside of the electric control cabinet, and inside the electric control cabinet, there is a multi-layer structure, and temperature sensors are installed at the two farthest positions of each layer.

[0064] At this point, the temperature collection is completed.

[0065] S002, obtaining the controlled temperature at each historical moment, and obtaining the first importance of the temperature sensor according to the correlation between each temperature sensor and all the temperatures of other temperature sensors, and the correlation between each temperature sensor and the controlled temperature at each historical moment.

[0066] The temperature values ​​collected by the sensor before the current moment are collectively referred to as historical data, and the temperature value collected at the current moment is called current data. By analyzing the similarity between historical data and current data, the adjustment results of the temperature control device corresponding to the historical data are used as the adjustment method of the temperature control device of the current data based on the obtained similarity.

[0067] Inside the electric control cabinet, the electric control cabinet must maintain a certain temperature range so that the components are not damaged. The temperature of the electric control cabinet is regulated by a temperature control device. The temperature control device also has a temperature range, so the temperature range is set. Each temperature value in the temperature range corresponds to a temperature level. The temperature level is the temperature value adjusted by the temperature control device. The temperature of the electric control cabinet is then adjusted by the temperature control device. Therefore, it is necessary to analyze the correlation between different temperature values ​​in the electric control cabinet and the regulated temperature of the temperature control device.

[0068] Preferably, in one embodiment of the present application, the temperature range of the temperature control device is [10, 30], wherein each temperature value is regarded as a temperature level, that is, 10-30 has a total of 21 temperature levels.

[0069] Preferably, in one embodiment of the present application, the temperature range of the temperature control device is [10, 25], wherein each temperature value is regarded as a temperature level, that is, 10-25 has a total of 16 temperature levels.

[0070] As for historical data, the control temperature corresponding to each moment can be obtained through the temperature level corresponding to each moment in history.

[0071] For historical data, the temperature values ​​corresponding to all temperature sensors at each moment constitute a multidimensional data point. The control temperature corresponding to each category is analyzed by clustering the multidimensional data points. However, since there are many temperature sensors, the correlation between the data collected by each temperature sensor and the control temperature is different. If the data in the temperature sensor is not related to the control temperature, it means that the temperature sensor collects invalid data, so it needs to be eliminated to eliminate its impact on the clustering.

[0072] Since the temperature change of the electric control cabinet is affected by the temperature at different positions, the present application analyzes the temperature information of the electric control cabinet in the obtained historical data, thereby obtaining the correlation between the temperature of the electric control cabinet and the temperature adjustment method of the electric control cabinet. And according to the correlation analysis, the influence of the temperature at different positions in the electric control cabinet on the temperature adjustment result of the electric control cabinet is obtained.

[0073] For the analysis of historical data, for each temperature sensor, all the temperature values ​​collected by it are sequenced in time order and recorded as a temperature sequence, and the correlation value between each temperature sequence and the other temperature sequences is calculated and recorded as the first correlation value; all the control temperatures corresponding to the historical data are sequenced in time order and recorded as a control sequence, and the correlation value between each temperature sequence and the control sequence is calculated and recorded as the second correlation value. The method for calculating the correlation value is the Pearson correlation coefficient between the sequences.

[0074] It should be noted that the Pearson correlation coefficient in this embodiment is only used to calculate the correlation value between two sequences, and the implementer may also use other correlation calculation methods, such as the Spearman rank correlation coefficient.

[0075] The importance factor of each temperature sensor is obtained according to the correlation values ​​between the temperature sequences and the correlation values ​​between the temperature sequences and the control sequences, and the importance factor of the temperature sensor is recorded as the first importance.

[0076] Any temperature sequence is recorded as a target sequence, and the importance factor of each temperature sensor is positively correlated with the first correlation value between the target sequence and the remaining temperature sequences and the second correlation value corresponding to the target sequence; the importance factor of each temperature sensor is negatively correlated with the second correlation value corresponding to the remaining temperature sequences and the difference between the second correlation values ​​of the target sequence and the remaining temperature sequences.

[0077] Preferably, in one embodiment of the present application, the product of the first correlation value between the target sequence and each temperature sequence and the second correlation value corresponding to the target sequence is used as the temperature importance; the absolute value of the difference between the target sequence and the second correlation value corresponding to each temperature sequence and the product of the second correlation value corresponding to each temperature sequence are recorded as the temperature influence value; the ratio of the temperature importance and the temperature influence value of the target sequence to each temperature sequence is accumulated and normalized as the importance factor of the temperature sensor corresponding to the target sequence.

[0078] Preferably, in one embodiment of the present application, the sum of the first correlation value between the target sequence and each temperature sequence and the second correlation value corresponding to the target sequence is used as the temperature importance; the absolute value of the difference between the target sequence and the second correlation value corresponding to each temperature sequence and the sum of the second correlation value corresponding to each temperature sequence are recorded as the temperature influence value; the ratio of the temperature importance and the temperature influence value of the target sequence to each temperature sequence is accumulated and normalized as the importance factor of the temperature sensor corresponding to the target sequence.

[0079] It should be noted that positive correlation means that when one variable increases, the other variable also increases, and the two variables change in the same direction. When one variable changes from large to small or from small to large, the other variable also changes from large to small or from small to large. The specific relationship can be a multiplication relationship, an addition relationship, or the idempotence of an exponential function, which is determined by actual application and is not specifically limited in this application.

[0080] It should be noted that negative correlation means that as one variable increases, the other variable decreases accordingly, and the two variables change in opposite directions. When one variable changes from large to small or from small to large, the other variable also changes from small to large or from large to small. The specific relationship can be a ratio relationship, a subtraction relationship, etc., which is determined by actual application, and this application does not impose any special restrictions.

[0081] So far, the first importance of each temperature sensor has been obtained.

[0082] S003, detecting abnormal points of all temperatures of the temperature sensor, and determining the second importance of the temperature sensor according to the time difference of all abnormal points of the temperature sensor and the temperature difference between the temperature sensor and other temperature sensors at the same time.

[0083] If a temperature sensor has more data exceeding the temperature range of the electric control cabinet in its historical data, it means that the electronic components at the location of the temperature sensor are more likely to be damaged, that is, the location of the temperature sensor needs to be temperature controlled to reduce damage. Therefore, the temperature sensor cannot be screened out, that is, the temperature sensor is more important.

[0084] Inside the electric control cabinet, the temperature in the temperature sensor that exceeds the temperature range of the electric control cabinet is taken as abnormal data, and the abnormal points in the temperature sequence corresponding to each temperature sensor are detected by an abnormal detection algorithm, and all abnormal points in all temperature sequences are counted. In this embodiment, the abnormal detection algorithm used is the LOF abnormal detection algorithm.

[0085] It should be noted that the LOF anomaly detection algorithm in this embodiment is only used to detect the abnormal points of each temperature sequence, and the implementer may also use other anomaly detection algorithms to obtain, for example, outlier detection.

[0086] For each abnormal point, the mean of the time intervals between the abnormal point and its two adjacent abnormal points is taken as the time difference of the abnormal point. For the first and last abnormal points, only the time interval of one adjacent abnormal point is calculated. The time differences of all abnormal points are constructed into a sequence recorded as a time regularity sequence.

[0087] Calculate the variance of all time differences in the time regularity sequence, and use the variance as the time regularity value of the time regularity sequence. The larger the value, the smaller the variance, that is, the closer the time differences between all abnormal points, that is, the closer the time distance between the abnormal points, that is, the more regular the abnormal points appear. The larger the time regularity value, the more the temperature change trend of the entire temperature sequence shows a periodic fluctuation, which means that the abnormal points appear regularly in the subsequent period, that is, the temperature sensor is more abnormal, that is, more important.

[0088] For the target sequence, the absolute value of the temperature difference between the target sequence and each of the other temperature sequences at each moment is calculated to form a temperature difference sequence between the target sequence and each temperature sequence. The variance of the absolute value of all temperature differences in the temperature difference sequence is calculated, and the inverse of the variance is recorded as the temperature difference change abnormality factor. The average value of the temperature difference change abnormality factor between the target sequence and all temperature sequences is taken as the temperature difference change abnormality.

[0089] By calculating the difference between the temperature sequences of each temperature sensor and the remaining temperature sensors, if the difference between the temperature sensor and the remaining temperature sensors is the same, that is, the variance is small, it means that the remaining temperature sensors have similar temperature changes, that is, the calculated temperature sensor has a large difference in temperature with the remaining temperature sensors, so the temperature sensor is more important.

[0090] The abnormal importance of the temperature sensor is determined according to the time regularity value of the temperature sensor and the abnormal degree of the temperature difference change, and the abnormal importance of the temperature sensor is recorded as the second importance.

[0091] The importance of anomaly is positively correlated with the time regularity value of the temperature sensor and the abnormality of the temperature difference change of the temperature sensor.

[0092] Preferably, in one embodiment of the present application, the time regularity value of the temperature sensor and the abnormality of the temperature difference change of the temperature sensor are multiplied together as the abnormal importance of the temperature sensor.

[0093] Preferably, in one embodiment of the present application, an exponential function is constructed with the abnormal degree of temperature difference change of the temperature sensor as the base and the time regularity value of the temperature sensor as the exponent, and the result of the exponential function is the abnormal importance of the temperature sensor.

[0094] At this point, the second importance of each temperature sensor is obtained.

[0095] S004: Determine the third importance of the temperature sensor according to the significance of all temperatures of each temperature sensor and the temperature difference between the temperature sensor and other surrounding temperature sensors.

[0096] Inside the electric control cabinet, since temperature control is more likely to be regulated due to higher temperatures and ignore the information of lower temperature areas, it is necessary to analyze the temperature information collected based on historical data and the temperature information obtained from different temperature sensor positions in the electric control cabinet, linearly normalize all temperature values ​​of each temperature sensor, and use the normalized temperature value as the significance of each temperature sensor at the current temperature value. Then, the significance of all temperature values ​​in the temperature sensor is averaged as the temperature significance value of the temperature sensor.

[0097] It should be noted that linear normalization is only one method of normalizing temperature values, and implementers may also use other normalization methods, such as maximum and minimum value normalization.

[0098] Among them, the higher the temperature significance value of the temperature sensor is, the higher the temperature collected by the temperature sensor as a whole is, and the more important the temperature sensor is.

[0099] Since the temperature difference of different temperature sensors in the electric control cabinet is different, if the temperature difference between one temperature sensor and the temperature sensors around it is larger, it means that the location of the temperature sensor is more likely to have a temperature out of range.

[0100] Taking the front of the electric control cabinet as the reference, for each temperature sensor, a two-dimensional coordinate system is constructed with it as the center. The front of the electric control cabinet is parallel to the two-dimensional coordinate system, and the temperature sensor at the center of the two-dimensional coordinate system is recorded as the center sensor. For the center sensor, all other temperature sensors are divided into four quadrants in the two-dimensional coordinate system, where 0 degrees is in the first quadrant, 90 degrees is in the second quadrant, 180 degrees is in the third quadrant, and 270 degrees is in the fourth quadrant.

[0101] The distance between each temperature sensor and the center sensor is calculated. In this embodiment, the distance is represented by the Euclidean distance between the center sensor and the temperature sensor in the two-dimensional coordinate system. The temperature sensors closest to the center sensor are selected in each of the four quadrants and recorded as neighboring sensors. If there is no temperature sensor in a quadrant, no temperature sensor is selected.

[0102] The degree of temperature change of the central sensor is determined based on the difference between all temperature values ​​of the central sensor and neighboring sensors.

[0103] The degree of temperature variation is positively correlated with the difference between all temperature values ​​of the central sensor and each neighboring sensor.

[0104] Preferably, in one embodiment of the present application, for each central sensor, the neighboring sensors in its four quadrants are obtained, the absolute value of the temperature difference between the neighboring sensors and the central sensor at each moment is calculated, and the sum of the absolute values ​​of the temperature difference at all moments is linearly normalized as the temperature change degree of the central sensor.

[0105] Preferably, in one embodiment of the present application, for each central sensor, the neighboring sensors within its four quadrants are obtained, the absolute value of the temperature difference between the neighboring sensors and the central sensor at each moment is calculated, and the product of the absolute values ​​of the temperature difference at all moments is normalized by the maximum and minimum values ​​and used as the temperature change degree of the central sensor.

[0106] The temperature distribution importance of the temperature sensor is obtained according to the temperature variation degree and the temperature significance value of each temperature sensor, and the temperature distribution importance of the temperature sensor is recorded as the third importance.

[0107] The importance of temperature distribution of temperature sensors is positively correlated with the degree of temperature change and the temperature significance value.

[0108] Preferably, in one embodiment of the present application, the product of the temperature variation degree of the temperature sensor and the temperature significance value is used as the temperature distribution importance of the temperature sensor.

[0109] Preferably, in one embodiment of the present application, the sum of the temperature variation degree and the temperature significance value of the temperature sensor is used as the temperature distribution importance of the temperature sensor.

[0110] So far, the third importance of each temperature sensor has been obtained.

[0111] S005, forwardly fuse the first importance, the second importance, and the third importance of the temperature sensor to obtain the importance of the temperature sensor inside the electric control cabinet.

[0112] According to the above analysis, the importance factor of each temperature sensor, the abnormal importance of the temperature sensor and the temperature distribution importance of the temperature sensor can be obtained to obtain the internal importance of the temperature sensor.

[0113] The three importances obtained for each temperature sensor are calculated based on the characteristics between the temperature sensors inside the control cabinet.

[0114] Therefore, the internal importance of the electric control cabinet is constructed according to the importance factor of each temperature sensor, the abnormal importance and the temperature distribution importance of the temperature sensor, that is, the internal importance of each temperature sensor is obtained according to the importance factor, the abnormal importance and the temperature distribution importance of the temperature sensor.

[0115] The internal importance of the temperature sensor is positively correlated with the importance factor, anomaly importance, and temperature distribution importance of the temperature sensor, respectively.

[0116] Preferably, in one embodiment of the present application, the product of the importance factor, the abnormal importance, and the temperature distribution importance of the temperature sensor is normalized and used as the internal importance of the temperature sensor.

[0117] Preferably, in one embodiment of the present application, the sum of the importance factor, the abnormal importance, and the temperature distribution importance of the temperature sensor is normalized and used as the internal importance of the temperature sensor.

[0118] So far, the internal importance of each temperature sensor inside the electric control cabinet has been obtained.

[0119] S006, obtaining the importance of the temperature sensor outside the electric control cabinet according to the difference between all temperatures collected by the temperature sensors outside the electric control cabinet and inside the electric control cabinet and the correlation of all controlled temperatures.

[0120] Since the purpose of adjusting the electrical control cabinet is mainly to ensure the stability of the internal temperature of the electrical control cabinet, the main monitoring target of the control is still the internal temperature information of the electrical control cabinet. However, the external environmental information will also have a certain impact on the temperature control results. Therefore, this application analyzes the importance of external temperature parameters based on the analysis results of the internal and external environmental temperature differences and the temperature control results.

[0121] Inside the electric control cabinet, for the temperatures of all temperature sensors at each moment, the Otsu threshold algorithm is used to divide the temperature values ​​into two parts, where all temperature values ​​in the part with a higher average temperature value constitute a high-temperature sequence, and the temperature of any temperature sensor outside the electric control cabinet at the current moment is subtracted from each temperature in the high-temperature sequence. The average of all the differences at each moment is taken as the internal and external temperature difference of the electric control cabinet at each moment, and the sequence formed by the internal and external temperature differences of each temperature sensor outside the electric control cabinet at all moments is recorded as the internal and external temperature difference sequence.

[0122] It should be noted that the role of the Otsu threshold algorithm here is only to classify the temperature values ​​into high temperatures and low temperatures. The implementer can also use other methods to classify the temperatures, such as directly presetting the threshold.

[0123] The correlation value between the calculated temperature internal and external difference sequence and the control sequence is recorded as the third correlation value, and the third correlation value is used as the external importance of the external temperature sensor of the electric control cabinet.

[0124] At this point, the external importance of each temperature sensor outside the electric control cabinet is obtained.

[0125] S007, selecting a temperature sensor with high importance according to the difference in correlation between the temperature values ​​of the temperature sensor inside the electric control cabinet and the temperature sensor outside the electric control cabinet and the control temperature.

[0126] Through the above steps, the internal importance of each temperature sensor inside the electric control cabinet and the external importance of each temperature sensor outside the electric control cabinet can be obtained. Since the main control purpose of the electric control cabinet is to control the temperature inside the electric control cabinet, the importance of different temperature sensors outside the electric control cabinet is suppressed to highlight the importance of the temperature sensor inside the electric control cabinet.

[0127] For each temperature sensor outside the electric control cabinet, obtain the temperature sequence corresponding to the temperature sensor, calculate the correlation value between the temperature sequence and the control sequence and record it as external correlation. The larger the correlation value, the more important the temperature sensor outside the electric control cabinet is.

[0128] At the same time, for the inside of the electric control cabinet, the correlation value between the temperature sequence of each temperature sensor in the electric control cabinet and the control sequence is divided into two categories using the Otsu threshold method, and the mean of the correlation value of each category is calculated. The mean of the correlation value corresponding to the class with the largest mean of the correlation value is recorded as the internal correlation. The larger the correlation, the weaker the external information is and the less important it is.

[0129] The inhibition factor is obtained through internal correlation and external correlation, and the ratio of external correlation to internal correlation is used as the inhibition factor. When the internal correlation is larger, it means that the external temperature sensor is less important and the inhibition factor is larger. When the external correlation is larger, it means that the external temperature sensor is more important and the inhibition factor is smaller.

[0130] The product of the suppression factor and the external importance of the temperature sensor outside the electric control cabinet is normalized as the importance screening value, and the internal importance of the temperature sensor inside the electric control cabinet is used as the importance screening value.

[0131] A screening threshold is preset, and temperature sensors with importance screening values ​​greater than the screening threshold are regarded as important sensors. In this embodiment, the screening threshold is 0.7, and the implementer can set it by himself.

[0132] At this point, the screening of temperature sensors is completed.

[0133] S008, determining a control temperature through a temperature sensor of high importance, and controlling the temperature of the device to a value corresponding to the control temperature.

[0134] For each important sensor, the current temperature and the historical temperature are clustered. The temperature level is the number of clustering categories. Each category corresponds to a control temperature. The control temperature corresponding to each important sensor at the current moment is obtained and recorded as the real-time control temperature.

[0135] Preferably, in one embodiment of the present application, the clustering algorithm used is the FCM clustering algorithm, the number of cluster categories is the number of temperature levels, and the cluster distance is the Euclidean distance between two temperature values.

[0136] For all important sensors, a real-time control temperature is obtained. Each real-time control temperature corresponds to a number. The real-time control temperature with the most corresponding numbers is used as the final control temperature. If there are multiple real-time control temperatures with the same number, the number of clustered temperature values ​​corresponding to each real-time control temperature is added together, and the real-time control temperature with the most temperature values ​​is used as the final control temperature. The temperature of the temperature control equipment is adjusted to the final control temperature to complete the control of the electric control cabinet. All steps of the process to complete the control of the electric control cabinet are as follows: Figure 2 shown.

[0137] It should be noted that the above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

[0138] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. The temperature control method of electric control cabinet based on artificial intelligence is characterized in that: The method comprises the following steps: Temperature collection steps: collect the temperature of temperature sensors at different locations at each moment; Steps to determine the importance of the temperature sensor inside the electric control cabinet: Obtain the controlled temperature at each historical moment, and obtain the first importance of the temperature sensor according to the correlation between each temperature sensor and all the temperatures of other temperature sensors, and the correlation between each temperature sensor and the controlled temperature at each historical moment; Detecting abnormal points of all temperatures of the temperature sensor, and determining the second importance of the temperature sensor according to the time difference of all abnormal points of the temperature sensor and the temperature difference of the temperature sensor and other temperature sensors at the same time; Determining the third importance of the temperature sensor based on the significance of all the temperatures of each temperature sensor and the temperature difference between the temperature sensor and the rest of the surrounding temperature sensors; The first importance, second importance, and third importance of the temperature sensor are forwardly integrated to obtain the importance of the temperature sensor inside the electric control cabinet; Steps to determine the importance of the external temperature sensor of the electric control cabinet: The importance of the temperature sensor outside the electric control cabinet is obtained based on the difference between all temperatures collected by the temperature sensors outside the electric control cabinet and inside the electric control cabinet and the correlation of all controlled temperatures; Temperature sensor screening steps: Select the temperature sensor with high importance according to the difference in correlation between the temperature values ​​of the internal temperature sensor and the external temperature sensor of the electric control cabinet and the control temperature; Temperature control steps: determine the control temperature through a high-importance temperature sensor, and control the temperature of the device to a value corresponding to the control temperature.

2. The temperature control method of the electric control cabinet based on artificial intelligence according to claim 1, characterized in that: The method for obtaining the first importance of the temperature sensor according to the correlation between each temperature sensor and all the temperatures of the other temperature sensors and the correlation between each temperature sensor and the control temperature at the historical moment is: At the historical moment, for each temperature sensor, all the temperature values ​​collected by it are sequenced in time order and recorded as a temperature sequence, and the correlation value between each temperature sequence and the other temperature sequences is calculated and recorded as the first correlation value; all the control temperatures corresponding to the historical data are sequenced in time order and recorded as a control sequence, and the correlation value between each temperature sequence and the control sequence is calculated and recorded as the second correlation value; Obtaining the importance factor of the temperature sensor according to the correlation values ​​between the temperature sequences and the correlation values ​​between the temperature sequences and the control sequences, wherein the importance factor of the temperature sensor is the first importance; Any temperature sequence is recorded as a target sequence, and its corresponding temperature sensor is the target sensor. The importance factor of each target sensor is positively correlated with the first correlation value between the target sequence and the remaining temperature sequences and the second correlation value corresponding to the target sequence; the importance factor of each target sensor is negatively correlated with the second correlation value corresponding to the remaining temperature sequences and the difference between the second correlation values ​​of the target sequence and the remaining temperature sequences.

3. The temperature control method of the electric control cabinet based on artificial intelligence as claimed in claim 2 is characterized in that: The method for obtaining the importance factor of the temperature sensor according to the correlation values ​​between the temperature sequences and the correlation values ​​between the temperature sequences and the control sequences is: The product of the first correlation value between the target sequence and each temperature sequence and the second correlation value corresponding to the target sequence is taken as the temperature importance; The product of the absolute value of the difference between the target sequence and the second correlation value corresponding to each temperature sequence and the second correlation value corresponding to each temperature sequence is recorded as the temperature influence value. After accumulating the ratio of the temperature importance and temperature influence value of the target sequence and each temperature sequence, the normalized result is used as the importance factor of the temperature sensor corresponding to the target sequence.

4. The temperature control method of the electric control cabinet based on artificial intelligence according to claim 2, characterized in that: The method for determining the second importance of the temperature sensor according to the time difference of all abnormal points of the temperature sensor and the temperature difference of the temperature sensor and the other temperature sensors at the same time is: For each target sequence, calculate the time interval between each abnormal point and its adjacent abnormal points, take the mean of the time intervals as the time difference of each abnormal point, and construct a sequence of the time differences of all abnormal points as a time regularity sequence; Calculating the variance of all time differences in the time regularity sequence, and using the variance as the time regularity value of the time regularity sequence; For the target sequence, the abnormal degree of temperature difference change is obtained according to the temperature difference between the target sequence and each other temperature sequence at each moment; The abnormal importance of the temperature sensor is determined according to the time regularity value of the temperature sensor and the abnormal degree of the temperature difference change, and the abnormal importance of the temperature sensor is the second importance; The importance of anomaly is positively correlated with the time regularity value of the temperature sensor and the abnormality of the temperature difference change of the temperature sensor.

5. The temperature control method of the electric control cabinet based on artificial intelligence according to claim 4 is characterized in that: The method for obtaining the abnormal degree of temperature difference change based on the temperature difference at each moment between the target sequence and each of the remaining temperature sequences is: For the target sequence, the absolute value of the temperature difference between the target sequence and each of the other temperature sequences at each moment is calculated to form a temperature difference sequence between the target sequence and each temperature sequence. The variance of the absolute value of all temperature differences in the temperature difference sequence is calculated, and the inverse of the variance is recorded as the temperature difference change abnormality factor. The average value of the temperature difference change abnormality factor between the target sequence and all temperature sequences is taken as the temperature difference change abnormality.

6. The temperature control method of the electric control cabinet based on artificial intelligence as claimed in claim 4 is characterized in that: The method for determining the abnormal importance of the temperature sensor according to the time regularity value and the temperature difference change abnormality of the temperature sensor is: The abnormal importance of the temperature sensor is obtained by multiplying the time regularity value of the temperature sensor and the abnormality degree of the temperature difference change of the temperature sensor.

7. The temperature control method of the electric control cabinet based on artificial intelligence as claimed in claim 4 is characterized in that: The method for determining the third importance of the temperature sensor according to the significance of all temperatures of each temperature sensor and the temperature difference between the temperature sensor and the rest of the surrounding temperature sensors is: All temperature values ​​of each temperature sensor are linearly normalized as the significance corresponding to the temperature value, and the average of all significances of the temperature sensor is taken as the temperature significance value of the temperature sensor; Any temperature sensor is recorded as the central sensor, and a two-dimensional coordinate system is constructed with it as the center. All temperature sensors are divided into four quadrants. The temperature sensors closest to the central sensor are selected from the four quadrants respectively, and these temperature sensors are recorded as neighborhood sensors. Determine the degree of temperature change of the central sensor based on the difference between all temperature values ​​of the central sensor and neighboring sensors; The degree of temperature variation is positively correlated with the difference between all temperature values ​​of the central sensor and each neighboring sensor; Obtaining the temperature distribution importance of the temperature sensor according to the temperature variation degree and the temperature significance value of each temperature sensor, wherein the temperature distribution importance of the temperature sensor is the third importance of the temperature sensor; The importance of temperature distribution of temperature sensors is positively correlated with the degree of temperature change and the temperature significance value.

8. The temperature control method of the electric control cabinet based on artificial intelligence according to claim 7, characterized in that: The method for forward fusion of the first importance, the second importance, and the third importance of the temperature sensor to obtain the importance of the temperature sensor inside the electric control cabinet is: Obtain the internal importance of each temperature sensor according to the importance factor, abnormal importance, and temperature distribution importance of the temperature sensor; The internal importance of the temperature sensor is positively correlated with the importance factor, anomaly importance, and temperature distribution importance of the temperature sensor, respectively.

9. The temperature control method of an electric control cabinet based on artificial intelligence according to claim 1, characterized in that: The method for obtaining the importance of the temperature sensor outside the electric control cabinet according to the difference between all temperatures collected by the temperature sensors outside the electric control cabinet and inside the electric control cabinet and the correlation of all controlled temperatures is: Inside the electric control cabinet, all temperatures at each moment are divided into two parts, and all temperature values ​​in the part with the higher mean temperature value constitute a high temperature sequence. The temperature of any temperature sensor outside the electric control cabinet at the same moment is subtracted from each temperature in the high temperature sequence, and the mean of all differences at each moment is taken as the internal and external temperature difference of the electric control cabinet at each moment. The sequence formed by the internal and external temperature differences of each temperature sensor outside the electric control cabinet at all moments is recorded as the internal and external temperature difference sequence; The correlation value between the calculated temperature internal and external difference sequence and the control sequence is recorded as the third correlation value, and the third correlation value is used as the external importance of the external temperature sensor of the electric control cabinet.

10. The temperature control method of electric control cabinet based on artificial intelligence according to claim 1, characterized in that: The method for selecting a temperature sensor with high importance according to the difference in correlation between the temperature values ​​of the temperature sensor inside the electric control cabinet and the temperature sensor outside the electric control cabinet and the control temperature is: For each temperature sensor outside the electric control cabinet, obtain the temperature sequence corresponding to the temperature sensor, calculate the correlation value between the temperature sequence and the control sequence and record it as external correlation; For the interior of the electric control cabinet, the correlation value between the temperature sequence of each temperature sensor in the electric control cabinet and the control sequence is divided into two categories using the threshold segmentation method, and the mean of the correlation value of each category is calculated. The mean of the correlation value corresponding to the category with the larger mean of the correlation value is recorded as the internal correlation; The ratio of external correlation to internal correlation was used as the suppression factor; The product of the suppression factor and the external importance of the external temperature sensor of the electric control cabinet is normalized as the importance screening value; A screening threshold is preset, and temperature sensors whose importance screening values ​​are greater than the screening threshold are regarded as important sensors.

Citation Information

Patent Citations

  • Temperature control method for intelligent file cabinet

    CN116501108A

  • Communication machine room temperature control method and system

    CN117130409A