An intelligent analysis method for the operating efficiency of an air compressor thermal management system

Abstract

The present invention relates to the field of digital data processing, and the present invention relates to an intelligent analysis method for the operating efficiency of an air compressor thermal management system, which collects data of multiple dimensions during the operation of the air compressor thermal management system, divides it into multiple time periods, and calculates the abnormality correction value and local correlation of the data. Based on this, the target data is corrected, the performance deviation is calculated, and the comprehensive efficiency score of the system operation is obtained. Finally, if the comprehensive efficiency score exceeds a preset threshold, an abnormal notification is issued. The present invention accurately identifies and corrects noise data by utilizing multi-dimensional data collection and intelligent analysis methods, thereby improving the operating efficiency and reliability of the air compressor thermal management system.

Description

Technical Field

[0001] The invention relates to the field of digital data processing, and in particular to an intelligent analysis method for the operating efficiency of a thermal management system of an air compressor. Background Art

[0002] The air compressor thermal management system is an important component to ensure the stable operation of the air compressor. Its main function is to keep the air compressor running within the optimal temperature and pressure range by controlling the cooling water circulation and lubricating oil system. The system includes devices such as temperature sensors, pressure sensors and flow meters, which are used to monitor the operating status of the air compressor in real time to prevent damage caused by overheating or overpressure. With the development of industrial automation, the frequency and importance of air compressors in various application scenarios are increasing. Traditional thermal management systems often cannot fully reflect the operating status of the system. In addition, noisy data and measurement errors often lead to misjudgment of the system's operating status, increasing the difficulty of maintenance and operation. These problems are mainly caused by the limitations of data acquisition and analysis methods, as well as insufficient consideration of multi-parameter interactions.

[0003] The existing air compressor thermal management system has some shortcomings in solving the above problems. For example, the Chinese patent application document with publication number CN118128742A discloses a method that generates a physical health index of the air compressor by collecting the environmental data and physical performance data of the air compressor in the working state to monitor the equipment status of the air compressor when it is working. However, this method cannot accurately identify and correct noise data, resulting in a high misjudgment rate. In addition, the lack of analysis of multi-parameter interactions makes it impossible to comprehensively evaluate the operating status of the system, affecting the reliability and efficiency of the system. Summary of the invention

[0004] In view of the above-mentioned problem that noise data cannot be accurately identified and corrected, the present invention proposes an intelligent analysis method for the operating efficiency of an air compressor thermal management system, including: collecting data in multiple dimensions during the operation of the air compressor thermal management system, and the data of the same dimension at different sampling times constitute time series data; dividing each time series data into data segments corresponding to multiple time periods using a time window of a set length, and calculating an abnormality correction value of the target data; the abnormality correction value is positively correlated with the abnormality of the target data, and is positively correlated with the difference between the target data and the remaining data in the data segment to which it belongs; the target data refers to any data in any data segment; within a time window, the target data is calculated. The local correlation of the target data is positively correlated with the correlation of the remaining data in the multiple data segments except the specified data set; the specified data set refers to multiple data at the sampling time corresponding to the target data; the target data is corrected according to the abnormality correction value and the local correlation, and the performance deviation is calculated based on the difference between the corrected data segment and the rated data of the air compressor thermal management system to obtain the performance deviation of all data segments in the same time window; the average of the performance deviations of all data segments in the same time window is used as the comprehensive efficiency score of the air compressor thermal management system during operation; in response to the comprehensive efficiency score in at least one time window being greater than a preset threshold, an abnormality notification is issued.

[0005] The present invention can effectively identify and correct noise data by collecting data of multiple dimensions during the operation of the air compressor thermal management system, dividing the time periods, and calculating the abnormal degree correction value and local correlation of the data, thereby reducing the possibility of data misjudgment and improving the accuracy of the system operation data. By correcting the target data and calculating the performance deviation, the comprehensive efficiency score of the system operation is finally obtained. If the score exceeds the preset threshold, the system will issue an abnormality notification. Compared with the prior art, the method of the present invention can more comprehensively and accurately evaluate the operating status of the air compressor thermal management system, and improve the operating efficiency and reliability of the system.

[0006] Furthermore, the abnormality correction value is:

[0007] α l，i =arctan((|x l，i -μ 1，l |-|x l，i -μ 2，l |)×θ l，i )+2

[0008] where α l，i Indicates the abnormality correction value of the ith data in the lth data segment; μ 1,l represents the average value of the data except the i-th data in the l-th data segment; μ 2,l represents the average value of all data in the lth data segment; x l,irepresents the value of the i-th data in the l-th data segment; θ l,i Indicates the abnormality of the i-th data in the l-th data segment.

[0009] Furthermore, the local correlation is:

[0010]

[0011] Where P l(k),i Indicates the local correlation of the i-th data in the l-th data segment of the k-th dimension; Indicates the k-th dimension l-th data segment and the l-th data segment determined in the same time window The relevance of the lth data segment of the dimension; m represents the total number of dimensions; I represents the total amount of data in the lth data segment; x l(k),j represents the jth data in the lth data segment of the kth dimension; μ 1,l(k) It represents the average value of the remaining data except the i-th data in the l-th data segment of the k-th dimension; Indicates The jth data in the lth data segment of the dimension; Indicates The average value of the remaining data except the i-th data in the l-th data segment of the dimension.

[0012] By calculating the local correlation of the target data, the relationship between different data can be identified, the degree of data abnormality can be further verified, the accuracy of data correction can be improved, and the comprehensiveness and accuracy of system status assessment can be ensured.

[0013] Furthermore, the target data is corrected according to the abnormality correction value and the local correlation, and further includes:

[0014] x′ l(k),i =x l(k),i -(norm(P l(k),i ×α l(k),i )×(x l(k),i -μ 1,l(k) ));

[0015] In the formula, x′ l(k),i represents the corrected value of the ith data in the lth data segment of the kth dimension; x l(k),i represents the i-th data in the l-th data segment of the k-th dimension; P l(k),i Represents the local correlation of the i-th data in the l-th data segment of the k-th dimension; α l(k),i represents the abnormality correction value of the ith data in the lth data segment of the kth dimension; μ 1,l(k) It represents the average value of the remaining data except the i-th data in the l-th data segment of the k-th dimension; norm() is the maximum and minimum normalization function.

[0016] By combining the abnormality correction value and the local correlation to correct the target data, we can more comprehensively consider the correlation between data, reduce the possibility of misjudgment of single data, and improve the accuracy of the overall system operation status assessment.

[0017] Furthermore, collecting data in multiple dimensions during the operation of the air compressor thermal management system also includes: placing a temperature sensor and a pressure sensor at the air compressor outlet to collect temperature data and pressure data during the operation of the air compressor thermal management system; and collecting flow data through a built-in flow counter in the air compressor system.

[0018] By placing temperature and pressure sensors at the outlet of the air compressor to collect key parameter data and combining it with the flow data collected by the flow counter, multi-dimensional data collection can provide comprehensive system operation information, help accurately analyze the system status, identify abnormal conditions, and improve the system's monitoring accuracy and analysis accuracy.

[0019] Furthermore, each time series data is divided into data segments corresponding to multiple time periods using a time window of a set length, and also includes: the time window of the set length is 5 minutes.

[0020] Furthermore, the abnormality degree of the target data further includes: obtaining the abnormality degree of the target data using a Z-Score algorithm.

[0021] Furthermore, calculating the performance deviation based on the difference between the corrected data segment and the rated data of the air compressor thermal management system also includes: normalizing the difference between the target data in the corrected data segment and the rated data of the air compressor thermal management system to the maximum and minimum values; and calculating the average of all normalized differences in the data segment to obtain the performance deviation of the data segment.

[0022] By calculating the performance deviation based on the difference between the corrected data segment and the rated data, the actual operating status of the system can be more accurately reflected, performance anomalies can be discovered in a timely manner, the timeliness and effectiveness of system maintenance can be improved, and the stable operation of the system can be guaranteed.

[0023] The technical effects of the present invention are:

[0024] The present invention utilizes multi-parameter data acquisition and intelligent analysis methods to accurately identify and correct noise data, thereby improving the operating efficiency and reliability of the air compressor thermal management system. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] 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:

[0026] Figure 1 The figure schematically shows a flow chart of an intelligent analysis method for operating efficiency of an air compressor thermal management system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0027] 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.

[0028] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0029] An embodiment of an intelligent analysis method for the operating efficiency of an air compressor thermal management system:

[0030] like Figure 1 As shown, an intelligent analysis method for the operating efficiency of an air compressor thermal management system of the present invention includes:

[0031] S1. Collect the operating data of the air compressor thermal management system.

[0032] Monitor the actual temperature and pressure of the compressed air output by the air compressor and the flow parameter data in the cooling water circulation system in real time to obtain the rated temperature, pressure and flow parameter data of the air compressor itself. Place temperature sensors and pressure sensors at the outlet of the air compressor to collect temperature data and pressure data; collect cooling water circulation flow data through the built-in flow counter of the air compressor system; read the rated parameter data of the air compressor itself at one time through the air compressor manual document; in the present invention, it is preset to collect all the above parameter data once per second, and the implementers can adjust according to the actual situation, and transfer the collected time series data of different dimensions to the database. The dimensional data mentioned in the content of the present invention refers to the above-mentioned collected parameter data, which will not be explained one by one later.

[0033] Furthermore, the present invention also includes the following steps: preprocessing the collected raw data. The preprocessing step includes data cleaning, denoising and data standardization. Data cleaning is to remove obviously erroneous or invalid data records; denoising technology can use moving average filtering or wavelet transform method to reduce the impact of random noise; data standardization is to convert data of various dimensions into the same dimensional range to facilitate subsequent processing and analysis.

[0034] By placing temperature and pressure sensors at the outlet of the air compressor to collect key parameter data and combining it with the flow data collected by the flow counter, multi-dimensional data collection can provide comprehensive system operation information, help accurately analyze the system status, identify abnormal conditions, and improve the system's monitoring accuracy and analysis accuracy.

[0035] S2. Obtain the abnormality degree of the target data; obtain the abnormality degree correction value of the target data in the data segment.

[0036] In the process of using sensors to monitor air compressor equipment, there may be problems such as insufficient accuracy, measurement errors, electromagnetic interference, and data transmission errors that lead to noise data. If the current data is noise data, and if the noise data causes the average value to shift in a certain direction, then the data points originally distributed around the true central trend will appear more discrete relative to this distorted average value, resulting in unnecessary expansion of the standard deviation. This leads to a higher degree of abnormality in the calculation of some normal data. Therefore, the abnormality correction value of the target data in the data segment is calculated. If the value is larger, it means that the target data deviates more from the average level of other data in the data segment.

[0037] Specifically, the time series data of each dimension is segmented using a time window of a set length. The time window of set length in the present invention is 5 minutes, and the implementer can adjust it according to the actual situation. In one embodiment, the Z-Score algorithm can be used to obtain the abnormality degree of the target data. In another embodiment, the LOF outlier factor algorithm can be used to obtain the abnormality degree of the target data.

[0038] Calculate the abnormality correction value of the data. The specific calculation formula is as follows:

[0039] α l,i =arctan((|x l,i -μ 1,l |-|x l,i -μ 2,l |)×θ l,i )+2;

[0040] where α l,i Indicates the abnormality correction value of the ith data in the lth data segment; μ 1,l represents the average value of the data except the i-th data in the l-th data segment; μ 2,l represents the average value of all data in the lth data segment; x l,i represents the value of the i-th data in the l-th data segment; θ l,i Indicates the abnormality of the i-th data in the l-th data segment.

[0041] Among them, α l,iIt represents the correction value of the abnormality of the i-th data in the l-th data segment. The larger the value, the more the data deviates from the average level of other data in the data segment, and the more it deviates from the global average level, which means that the data is obviously abnormal relative to other data in its data segment. Therefore, the abnormality of the data causes the overall average level to be raised or lowered. On the contrary, if the value is smaller, it means that the data is approximately normal relative to other data in its data segment, and the abnormality of the data is considered to be lower; the arctan(x)+2 function can be used to map the above logic. When the j-th data in the l-th data segment is larger, the obtained α l,i The larger the value.

[0042] S3. Obtain the local correlation of the target data based on the correlation of the remaining data segments in the time window except for the multiple data corresponding to the target data at the corresponding sampling time.

[0043] Since data of different dimensions often interact with each other, for example, an increase in the internal pressure of the system may cause the exhaust port temperature to rise; the flow rate in the cooling water circulation directly affects the heat dissipation effect of the system, and thus affects the temperature of the air compressor.

[0044] Specifically, the local correlation of the data is calculated; the specific calculation formula is as follows:

[0045]

[0046] Where P l(k),i Indicates the local correlation of the i-th data in the l-th data segment of the k-th dimension; Indicates the k-th dimension l-th data segment and the l-th data segment determined in the same time window The relevance of the lth data segment of the dimension; m represents the total number of dimensions; I represents the total amount of data in the lth data segment; x l(k),j represents the jth data in the lth data segment of the kth dimension; μ 1,l(k) It represents the average value of the remaining data except the i-th data in the l-th data segment of the k-th dimension; Indicates The jth data in the lth data segment of the dimension; Indicates The average value of the remaining data except the i-th data in the l-th data segment of the dimension.

[0047] Among them, (x l(k),j -μ 1,l(k) ) represents the difference between the jth data except the i-th data in the l-th data segment of the k-th dimension and the mean of the remaining data except the i-th data in the data segment; and The larger the kth dimension is, the The trend of dimension changes is positively or negatively correlated. It can be considered that the higher the correlation between the two dimensions in the data segment corresponding to the same time window, the higher the correlation between the two dimensions. The larger the value, the greater the difference between the k-th dimension and the l-th data segment determined by the same time window in the data segment. The greater the correlation of the lth data segment in the dimension, the higher the local correlation of the i-th data in the lth data segment in the k-th dimension.

[0048] By calculating the local correlation of the target data, the relationship between different data can be identified, the degree of data abnormality can be further verified, the accuracy of data correction can be improved, and the comprehensiveness and accuracy of system status assessment can be ensured.

[0049] S4. Obtain a target data correction value based on the above abnormality correction value and local correlation.

[0050] If the abnormality correction value of a data is high, and the correlation between the time series data of other dimensions is significant without the influence of this data, it means that there is a close interaction between these dimensions under the normal operation of the system, and it can be shown that the data is more likely to be noise data. By correcting the mean deviation of the above data, non-systematic noise can be eliminated and data that more truly reflects the system status can be retained.

[0051] Specifically, for a certain data, the deviation of the data from the mean of other data in its data segment is corrected according to the abnormal degree correction value of the data and the local correlation degree, and the corrected value of the data is obtained. The specific calculation formula is as follows:

[0052] x′ l(k),i =x l(k),i -(norm(P l(k),i ×α l(k),i )×(x l(k),i -μ 1,l(k) ))

[0053] In the formula, x′ l(k),i represents the corrected value of the ith data in the lth data segment of the kth dimension; x l(k),i represents the i-th data in the l-th data segment of the k-th dimension; P l(k),i Represents the local correlation of the i-th data in the l-th data segment of the k-th dimension; α l(k),i represents the abnormality correction value of the ith data in the lth data segment of the kth dimension; μ 1,l(k) It represents the average value of the remaining data except the i-th data in the l-th data segment of the k-th dimension; norm() is the maximum and minimum normalization function.

[0054] Among them, (P l(k),i ×α l(k),i) is larger, the higher the abnormality correction value of the ith data in the lth data segment of the kth dimension is, and when there is no ith data in the lth data segment of the kth dimension, the higher the local correlation between the data in other dimensions is, the more likely the data is real noise data, and the greater the degree of correction should be. (x l(k),i -μ 1,l(k) ) represents the deviation between the ith data in the lth data segment of the kth dimension and the mean of the rest of the data in the lth data segment of the kth dimension except for this data. If the deviation is greater than 0 and the larger the deviation is, the greater the data is than the average value in the data segment, and the greater the correction is needed to reduce the data value. On the contrary, if the deviation is less than 0 and the larger the absolute value of the deviation is, the lower the data is than the average value in the data segment, and the greater the correction is needed to increase the data value. In summary, (P l(k),i ×α l(k),i )×(x l(k),i -μ 1,l(k) ) represents the corrected deviation value of the ith data in the lth data segment of the kth dimension. If the corrected deviation value is greater than 0 and the larger the value is, the corrected value of the ith data should be reduced. The corrected value of the ith data can be obtained by subtracting the corrected deviation value from the actual value of the data.

[0055] By combining the abnormality correction value and the local correlation to correct the target data, we can more comprehensively consider the correlation between data, reduce the possibility of misjudgment of single data, and improve the accuracy of the overall system operation status assessment.

[0056] S5. Obtain the performance deviation of the data segment based on the target data correction value and the rated data; obtain the comprehensive efficiency score of the thermal management system operation and implement abnormality notification based on the comprehensive efficiency score.

[0057] Obtain the corrected value of each data in each dimension, and combine the difference between the corrected value and the rated data to evaluate whether the system operation deviates from expectations. Once it exceeds expectations, the system's operating efficiency will decrease, and in severe cases, it may cause equipment failure. However, the overall performance of the air compressor operation process is often not determined by a single dimension, but the result of the interaction of multiple dimensions. By comprehensively evaluating the deviations in different dimensions, we can have a more comprehensive understanding of the system's operating status and efficiency level.

[0058] Specifically, the difference between the corrected value of the target data and the rated data of the dimension where the target data is located is calculated; then the mean of the difference between the corrected value of all data in the data segment where it is located and the rated data is calculated, and normalized, and it is used as the average deviation of the data segment in this dimension; further the average deviation of the data segments of the remaining dimensions in the time window corresponding to the data segment is calculated, and the mean of the average deviation of the data segments of all dimensions in the time window is used as the comprehensive efficiency score of the air compressor thermal management system in the time window, and the comprehensive efficiency score of the time window is recorded as S. The preset operating efficiency warning threshold T, if any time window γ has a comprehensive efficiency score S γ If it is greater than T, it means that the deviation of the air compressor operating data within the time window exceeds the preset normal range, and a notification of abnormal operation of the system air compressor thermal management system is issued and fed back to the maintenance personnel for timely maintenance.

[0059] In the present invention, the empirical value T=0.65 is taken, and the implementer can adjust it according to the actual situation.

[0060] By calculating the performance deviation based on the difference between the corrected data segment and the rated data, the actual operating status of the system can be more accurately reflected, performance anomalies can be discovered in a timely manner, the timeliness and effectiveness of system maintenance can be improved, and the stable operation of the system can be guaranteed.

[0061] The present invention can effectively identify and correct noise data by collecting data of multiple dimensions during the operation of the air compressor thermal management system, dividing the time periods, and calculating the abnormal degree correction value and local correlation of the data, thereby reducing the possibility of data misjudgment and improving the accuracy of the system operation data. By correcting the target data and calculating the performance deviation, the comprehensive efficiency score of the system operation is finally obtained. If the score exceeds the preset threshold, the system will issue an abnormality notification. Compared with the prior art, the method of the present invention can more comprehensively and accurately evaluate the operating status of the air compressor thermal management system, and improve the operating efficiency and reliability of the system.

[0062] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly defined.

[0063] 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. An intelligent analysis method for the operating efficiency of an air compressor thermal management system, characterized in that: The method comprises: Collect data in multiple dimensions during the operation of the air compressor thermal management system. The data of the same dimension at different sampling times constitutes time series data. Each time series data is divided into data segments corresponding to multiple time periods using a time window of a set length, and an abnormality correction value of the target data is calculated; the abnormality correction value is positively correlated with the abnormality of the target data, and is positively correlated with the difference between the target data and the remaining data in the data segment to which it belongs; the target data refers to any data in any data segment; In a time window, the local correlation of the target data is calculated, and the local correlation is positively correlated with the correlation of the remaining data in the multiple data segments except the specified data set; the specified data set refers to multiple data at the sampling time corresponding to the target data; The target data is corrected according to the abnormality correction value and the local correlation, and the performance deviation is calculated based on the difference between the corrected data segment and the rated data of the air compressor thermal management system to obtain the performance deviation of all data segments in the same time window; the average of the performance deviations of all data segments in the same time window is used as the comprehensive efficiency score of the air compressor thermal management system during operation; In response to the comprehensive efficiency score in at least one time window being greater than a preset threshold, an abnormal notification is issued.

2. The intelligent analysis method for the operating efficiency of an air compressor thermal management system according to claim 1 is characterized in that: The abnormality correction value is: a l，i =arctan((|x l，i -m 1，l |-|x l，i -m 2，l |)×θ l，i )+2; where α l,i Indicates the abnormality correction value of the i-th data in the l-th data segment; μ 1,l It represents the average value of the data except the i-th data in the l-th data segment; μ2 ,l represents the average value of all data in the lth data segment; x l,i represents the value of the i-th data in the l-th data segment; θ l,i Indicates the abnormality of the i-th data in the l-th data segment.

3. The intelligent analysis method for the operating efficiency of an air compressor thermal management system according to claim 1 is characterized in that: The local correlation is: Where P l(k),i Indicates the local correlation of the i-th data in the l-th data segment of the k-th dimension; Indicates the k-th dimension l-th data segment and the l-th data segment determined in the same time window The relevance of the lth data segment of the dimension; m represents the total number of dimensions; I represents the total amount of data in the lth data segment; x l(k),j represents the jth data in the lth data segment of the kth dimension; μ 1,l(k) It represents the average value of the remaining data except the i-th data in the l-th data segment of the k-th dimension; Indicates The jth data in the lth data segment of the dimension; Indicates The average value of the remaining data except the i-th data in the l-th data segment of the dimension.

4. The intelligent analysis method for the operating efficiency of an air compressor thermal management system according to claim 1 is characterized in that: Correcting the target data according to the abnormality correction value and the local correlation degree includes: x′ l(k),i =x l(k)，i -(norm(P l(k),i ×a l(k)，i )×(x l(k),i -m 1,l(k) )); In the formula, x′ l(k),i represents the corrected value of the ith data in the lth data segment of the kth dimension; x l(k),i represents the i-th data in the l-th data segment of the k-th dimension; P l(k),i Represents the local correlation of the i-th data in the l-th data segment of the k-th dimension; α l(k),i represents the abnormality correction value of the ith data in the lth data segment of the kth dimension; μ 1,l(k) It represents the average value of the remaining data except the i-th data in the l-th data segment of the k-th dimension; norm() is the maximum and minimum normalization function.

5. The intelligent analysis method for the operating efficiency of an air compressor thermal management system according to claim 1, characterized in that: Collect data from multiple dimensions during the operation of the air compressor thermal management system, including: Temperature sensors and pressure sensors are placed at the outlet of the air compressor to collect temperature data and pressure data during the operation of the air compressor thermal management system; flow data is collected through the built-in flow counter of the air compressor system.

6. The intelligent analysis method for the operating efficiency of an air compressor thermal management system according to claim 1, characterized in that: Each time series data is divided into data segments corresponding to multiple time periods using a time window of a set length, including: The time window of the set length is 5 minutes.

7. The intelligent analysis method for the operating efficiency of an air compressor thermal management system according to claim 1, characterized in that: The degree of abnormality of the target data, including: Use the Z-Score algorithm to obtain the abnormality degree of the target data.

8. The intelligent analysis method for the operating efficiency of an air compressor thermal management system according to claim 1, characterized in that: The performance deviation is calculated based on the difference between the corrected data segment and the rated data of the air compressor thermal management system, including: The difference between the data in the corrected data segment and the rated data of the air compressor thermal management system is normalized to the maximum and minimum values; The performance deviation of the data segment is obtained by calculating the average of all normalized differences in the data segment.

Images