Pressure early warning method and system based on threshold and state trend analysis

Abstract

The invention discloses a pressure early warning method and system based on threshold and state trend analysis, and the method comprises the steps: obtaining pressure time sequence data of a recently set first time period from a real-time gas database; calculating by using a sliding window moving average method to obtain average values of the current moment in the latest third time period, the latest fourth time period and the latest fifth time period; identifying a rising or falling trend according to the calculated mean value, and counting the duration of the trend; constructing a function relation between the trend target value and the duration; and according to an actual service scene rule, setting a threshold interval in a segmented manner, and judging whether to trigger an early warning mechanism or not according to a relationship between a real-time value and a threshold.

Description

Technical Field

[0001] The present application relates to the technical field of blast furnace gas pipeline network pressure control, and in particular, to a pressure warning method and system capable of improving the timeliness and accuracy of pressure warning. Background Art

[0002] Pressure warning monitoring is the most important part of blast furnace gas pipeline monitoring. Gas dispatchers need to pay attention to changes in pipeline pressure at all times. When the pipeline pressure is too low or too high, they need to notify the corresponding units to make production adjustments in order to maintain the pipeline pressure within an appropriate range. At present, there are few related contents in the field of blast furnace gas pressure warning research, and the only research applications only consider the impact of threshold factors.

[0003] The pressure warning method in the prior art is mainly based on the pressure signal data collected by the sensor, and is analyzed and processed through threshold judgment conditions to determine whether the pressure during detection exceeds a preset pressure threshold.

[0004] These methods have some limitations: (1) When processing a large amount of sensor data, only one value is retained per minute to avoid the problem of high computational complexity that may be faced when processing a large amount of data. This method leads to poor real-time performance and loss of trend information within minutes; (2) When judging whether the threshold is exceeded, only the real-time value is used as the basis for judgment, which may lead to misjudgment or warning delay for complex pressure change patterns. Summary of the invention

[0005] The purpose of this application is to provide a pressure warning method and system based on threshold and state trend analysis to improve the defects of the prior art such as poor real-time performance, loss of trend information within minutes, and possible misjudgment or warning delay.

[0006] The embodiments of the present application adopt the following technical solutions: A pressure early warning method based on threshold and state trend analysis, characterized in that it comprises the following steps: Step 1: Obtain pressure data, obtain the pressure time series data of the most recently set first time period from the real-time database, and perform outlier processing; for the processed data, take the average value and resample every second time period; Step 2: Use the sliding window moving average method to calculate the average of the current time in the third most recent time period, the fourth most recent time period, and the fifth most recent time period; Identify the upward or downward trend based on the calculated mean and count the duration of the trend; Identify the upward or downward trend based on the calculated mean, count the duration of the trend, and construct a functional relationship between the trend target value and duration; Step 3: Based on the functional relationship between the target value and the duration, calculate the time required to reach the target value at the current trend rate; according to the actual business scenario rules, set the threshold interval in segments, and determine whether to trigger the warning mechanism based on the relationship between the real-time value and the threshold.

[0007] In the above technical solution, in step one, outlier processing includes jump processing and missing value processing.

[0008] In the above technical solution, in step one, the jump processing is: compare the absolute value of the difference between the current value and the previous value, and the absolute value of the difference between the next value. If the two absolute values ​​differ too much at the same time, the pressure value at that moment is modified to the average of the two; missing values ​​and 0 values ​​are also filled with the average of the two jumps.

[0009] In the above technical solution, in step 1, the second time period of data resampling is much shorter than the first time period.

[0010] In the above technical solution, in step 2, in the moving average method of the sliding window, the relationship between the durations of the various time segments is that the third time period is smaller than the fourth time period, the fourth time period is smaller than the fifth time period; and the second time period is much smaller than the third time period.

[0011] Preferably, the first time period is in hours, the second time period is in seconds, and the third time period is in minutes.

[0012] In the above technical solution, in step 2, the rule configuration for identifying an upward trend or a downward trend is: if the current mean is greater than the previous mean, it is marked as rising; if the current mean is less than the previous mean, it is marked as falling.

[0013] In the above technical solution, in step 2, the trend target value is constructed y The functional relationship with the duration of the trend is: ; in x is the current real-time pressure value, y The target value set in advance, k is the pressure change rate of the two times before the current moment, is the pressure value 2 unit times before the current moment, and the unit time is the second time period; is the specific value of the duration or trend duration corresponding to x at the current moment, is the desired variable value.

[0014] In the above technical solution, in step 3, the relationship between the real-time value and the threshold value formed by setting the threshold interval in segments is configured as follows: Set the threshold interval [attention value lower limit, attention value upper limit, alarm value lower limit]. The thresholds in the threshold interval are arranged from small to large. When the real-time value is less than the attention value lower limit, the early warning mechanism is not triggered. When the attention value lower limit is less than the real-time value and less than the attention value upper limit, the early warning mechanism is triggered. When the real-time value is greater than the alarm value lower limit, the alarm information is pushed immediately.

[0015] The present invention also provides a pressure early warning system based on threshold value and state trend analysis, which is characterized by comprising: A data acquisition module is used to obtain pressure time series data from a real-time database; A data processing module, used for performing outlier processing and data resampling on the acquired data; the data processing module includes a moving average calculation function of a sliding window; The trend identification module identifies the upward or downward trend based on the calculated mean value, and counts the duration of the trend to construct a functional relationship between the trend target value and the duration; based on the functional relationship between the target value and the duration, the time required to reach the target value at the current trend rate is calculated; The early warning judgment module sets the threshold intervals in segments according to the actual business scenario rules, and determines whether to trigger the early warning mechanism based on the relationship between the real-time value and the threshold.

[0016] The present invention also provides a computer-readable storage medium, characterized in that a computer program is stored therein, and when the computer program is executed, it is used to implement the above-mentioned pressure warning method based on threshold value and state trend analysis.

[0017] In summary, the present invention provides a pressure trend warning method and system based on thresholds and state duration, aiming to solve the problem of false warnings in the prior art and improve the timeliness and accuracy of pressure warnings. In particular, since the pressure of the pipeline network is affected by production and consumption, the pressure change pattern is frequent and complex. Since the existing threshold warning model does not consider the pressure state, such as distinguishing the different effects of rising or falling trends, late reporting, false reporting, etc. may occur.

[0018] Compared with the prior art, the present invention has the following beneficial effects: The present invention considers the preset pressure threshold, pressure trend identification and pressure trend duration together to form a pressure early warning method based on threshold and state trend analysis, which includes data acquisition and preprocessing, trend state identification and statistics, and segmented threshold trend early warning algorithm. The system includes a data acquisition module, a data processing module, a trend identification module and an early warning judgment module. It can effectively improve the timeliness and accuracy of the pressure trend early warning model and make up for the lack and deficiency of the research on blast furnace gas pressure early warning method.

[0019] The present invention is not only applicable to trend warning of pipeline gas pressure, but also widely applicable to physical quantity warning scenarios related to thresholds and trend warnings, such as water pressure, voltage, etc.

[0020] The present invention aims to solve the problem of false warning in the prior art. By combining pressure trend identification and pressure trend duration, the present invention solves the problem that the prior art only retains one value per minute to avoid large data processing volume, resulting in poor real-time performance and easy loss of trend information within minutes.

[0021] By resampling the pressure to obtain the average value over a period of time, data fluctuations can be alleviated, and the pressure trend can be combined to further accurately determine whether an alarm is needed, thus solving the problems of false alarms and untimely alarms, making the warning results more accurate and reliable, and improving the timeliness and accuracy of pressure warnings.

[0022] Early warning: Based on the average pressure, trend, and duration, the alarm information is generated one and a half minutes earlier than the existing technology, giving users time to react and prepare measures, and further accurately judging whether an alarm is needed. And because of the comprehensive consideration of trend pressure balance, it can also effectively alleviate the impact of pressure fluctuations. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0024] Figure 1 This is the overall flow chart of Example 1 of the present application, including data acquisition, trend status identification and statistics, and segmented threshold trend warning. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.

[0026] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application for which protection is sought, but merely represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0027] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0028] In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, or the positions or positional relationships in which the product of the application is usually placed when in use. They are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific position, be constructed and operated in a specific position, and therefore cannot be understood as a limitation on the present application. In addition, the terms "first", "second", "third", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0029] In addition, the terms "horizontal", "vertical", "overhanging" and the like do not mean that the components are required to be absolutely horizontal or overhanging, but can be slightly tilted. For example, "horizontal" only means that its direction is more horizontal than "vertical", and does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0030] In the description of this application, it should also be noted that, unless otherwise clearly specified and limited, the terms "set", "install", "connect", and "connect" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two elements. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0031] In the present application, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may include that the first and second features are in direct contact, or may include that the first and second features are not in direct contact but are in contact through another feature between them. Moreover, a first feature being "above", "above" and "above" a second feature includes that the first feature is directly above and obliquely above the second feature, or simply indicates that the first feature is higher in level than the second feature. A first feature being "below", "below" and "below" a second feature includes that the first feature is directly below and obliquely below the second feature, or simply indicates that the first feature is lower in level than the second feature.

[0032] The features and performance of the present application are further described in detail below in conjunction with the embodiments.

[0033] Example 1 The pressure warning method based on threshold value and state trend analysis in Example 1 of the present application is as follows Figure 1 As shown, the following steps are included: Step 1: Obtain pressure data, obtain the pressure time series data of the most recently set first time period from the real-time database, and perform outlier processing; for the processed data, take the average value and resample every second time period; Step 2: Use the sliding window moving average method to calculate the average of the current time in the third most recent time period, the fourth most recent time period, and the fifth most recent time period; Identify the upward or downward trend based on the calculated mean and count the duration of the trend; Identify the upward or downward trend based on the calculated mean, count the duration of the trend, and construct a functional relationship between the trend target value and duration; Step 3: Based on the functional relationship between the target value and the duration, calculate the time required to reach the target value at the current trend rate; according to the actual business scenario rules, set the threshold interval in segments, and determine whether to trigger the warning mechanism based on the relationship between the real-time value and the threshold.

[0034] The specific steps are described as follows: Step 1: Pressure data acquisition and preprocessing 1.1 Data acquisition and outlier processing To analyze the pressure change trend, the pressure time series data (one pressure value every 2 seconds) of the last hour (set as the first time period) is obtained from the real-time database. Due to transmission communication problems, the original data may have jumps, missing values, or 0 values. Therefore, after obtaining the original data, preprocessing is performed for data anomalies in different scenarios: (1) Jump: Compare the absolute value of the difference between the current value and the previous value, and the absolute value of the difference between the next value. If the two absolute values ​​differ too much at the same time, the pressure value at that moment is modified to ; (2) Missing and 0 values: Use Filling. This method of processing abnormal data through linear interpolation can retain the trend of the original data.

[0035] 1.2 Data Resampling Since the time interval of the original pressure data is 2s, if the time interval is too small, two adjacent values ​​may be the same, resulting in errors in the trend change identification. Therefore, the pre-processed data is resampled to take the average value every 10s (set as the second time period) to achieve a balance between data interval accuracy and data availability, and improve the timeliness and accuracy of pressure data.

[0036] Step 2: Identification and statistics of rising and falling trend states 2.1 Mean calculation of different windows Moving average method using sliding window , where n is set to 6, 18, and 30 respectively, and the average of the current time within the last minute (set to the third time period), the last 3 minutes (set to the fourth time period), and the last 5 minutes (set to the fifth time period) is calculated. , , 2.2 Trend Identification Rules: If and , it rises and is marked as 1; like and , then it decreases and is marked as -1; Otherwise, mark it as 0.

[0037] According to the above calculation rules, the state at each time point is identified to obtain the state column at each time point, marked as S, such as [1,1,1,1,0,0,-1,-1,1,1,1,1,1,1].

[0038] 2.3 Trend duration statistics Set initial trend duration = 0.

[0039] Get the current state currentState. For example, if the current currentState=1, it is in the rising state. For the state column S obtained in 2.2, count from the back to the front. If the element is the same as the currentState, then "the trend duration + 1" is added. This continues until the first different element is encountered. Then the accumulation is terminated. For example, when currentState=1 and S=[1,1,1,1,0,0,-1,-1,1,1,1,1,1,1], the trend duration is calculated to be 6.

[0040] 2.4 Functional relationship between target value and duration Through steps 2.1-2.3, we can get: current real-time value x, trend status (up or down), and trend duration. Then we can construct the target value y and x , the functional relationship of the trend duration: ; Where y is the target value set in advance, x is the real-time value of pressure, k is the pressure change rate two times before the current moment, is the value (pressure value) of 2 unit time (the “unit time” is the second time period mentioned above) before the current moment. for x The corresponding duration (the specific value of the trend duration at the current moment), is the value of the variable to be sought. k It can be calculated by the duration of the trend corresponding to the two changes in the real-time value, and the time required to reach the target value can be calculated by the above formula.

[0041] For example, setting the target value to y =14, real-time value x =12, =11, is 7, and the trend duration to reach the set target value is . Then first find , and then substitute the known conditions into the formula You can get .

[0042] This means: According to the current rate of change, the target value will be reached when the duration changes to 11, which is a long way from now. Unit time.

[0043] Step 3: Segmented Threshold Trend Warning Algorithm 3.1 Interval Threshold Setting According to the actual business scenario rules, set the threshold interval [Note down, Note up, Report down], and arrange the thresholds from small to large. Note down represents the lower limit of the attention value, Note up represents the upper limit of the attention value, and Report down represents the lower limit of the alarm value. For the sake of simplicity, only the single-side high pressure warning problem is listed.

[0044] 3.2 Rules within different threshold ranges When the real-time value is less than the note, the early warning mechanism will not be triggered.

[0045] When the bet down < real-time value < bet up, the warning mechanism is triggered. First, identify the rising / falling trend through step 2.2. If it is in a falling trend, no alarm is given; If it is in an upward trend, the duration of the trend is further obtained through step 2.3. Only when the trend continues for a period of time will the warning information be pushed, and step 2.4 is used to give how long it will take for the pressure value to reach the alarm value based on the existing working conditions.

[0046] When the real-time value > the alarm value, the alarm information will be pushed immediately.

[0047] Example 2 The method of the present invention solves the problem of false alarms. Principle: 1. Resample the pressure to obtain the average value within a period of time to alleviate data fluctuations. 2. Under the condition of pressure discrimination, combine the pressure trend to further accurately determine whether an alarm is needed.

[0048] Table 1 Examples of solving false alarm problems

[0049] Table 1 shows the comparative effect of solving the false alarm problem by using Example 2 of the present invention compared with the existing simple threshold method.

[0050] From the data in the 7th and 8th columns shown in Table 1, it can be seen that the prior art predicts an alarm, while the alarm condition of this embodiment is: the pressure value is lower than 7. After analysis, it is confirmed that no alarm is needed.

[0051] In view of the above situation, if the existing threshold alarm model is used, the alarm information will be pushed. If the method of the present invention is used, no alarm will be displayed.

[0052] According to the pressure change after the threshold model alarm, it can be seen that the low pressure value at the alarm moment is caused by pressure fluctuations. After a short change, it returns to normal and no longer meets the alarm condition.

[0053] Therefore, the alarm of the threshold model can be considered as a false alarm. Using the method of the present invention, since the current value of the 7th column data after processing is 7.39 and the trend is in an upward state, no alarm is a correct judgment.

[0054] Example 3 By using the method of the present invention, an early warning is given.

[0055] Principle: 1. Judge based on mean pressure, trend and duration. Alarm condition in this embodiment: pressure value is lower than 7.

[0056] Table 2 below shows the comparative effect of early warning given by Example 3 of the present invention compared with the existing simple threshold method.

[0057] Table 2 Early warning examples

[0058] As shown in Table 2, if the existing simple threshold method is used, the first alarm condition is at 2024-8-12 8:25 and after the alarm, there is a missed alarm due to pressure fluctuations.

[0059] The first alarm using this method was at 8:23 on 2024-8-12. The alarm condition was that the 1-minute pressure average was less than 7, and the pressure trend continued to decline for a period of time. By combining this pressure and trend method, this embodiment can generate an alarm message 2 minutes in advance and avoid the impact of pressure fluctuations.

[0060] In other examples not shown, the alarm is given 1 minute 30 seconds or 1 minute 40 seconds in advance. It can be seen that the effect of giving early warning in the present invention is very obvious.

[0061] The embodiments described above are part of the embodiments of the present application, rather than all of the embodiments. The detailed description of the embodiments of the present application is not intended to limit the scope of the present application for protection, but merely represents the selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

Claims

1. A pressure early warning method based on threshold and state trend analysis, characterized in that: The following steps are involved: Step 1: Obtain pressure data, obtain the pressure time series data of the most recently set first time period from the real-time database, and perform abnormal value processing; For the processed data, the mean value is taken and resampled every second time period; Step 2: Use the sliding window moving average method to calculate the average of the current time in the third most recent time period, the fourth most recent time period, and the fifth most recent time period; Identify the upward or downward trend based on the calculated mean and count the duration of the trend; Identify the upward or downward trend based on the calculated mean, count the duration of the trend, and construct a functional relationship between the trend target value and duration; Step 3: Based on the functional relationship between the target value and the duration, calculate the time required to reach the target value at the current trend rate; According to the actual business scenario rules, the threshold intervals are set in segments, and the relationship between the real-time value and the threshold is used to determine whether to trigger the early warning mechanism.

2. The pressure early warning method based on threshold value and state trend analysis according to claim 1 is characterized in that: In step one, outlier processing includes jump processing and missing value processing.

3. The pressure early warning method based on threshold value and state trend analysis according to claim 1 is characterized in that: In step one, the jump processing is: compare the absolute value of the difference between the current value and the previous value, and the absolute value of the difference between the next value. If the two absolute values ​​differ too much at the same time, the pressure value at that moment is modified to the average of the two; missing values ​​and 0 values ​​are also filled with the average of the two jumps.

4. The stress early warning method based on threshold and state trend analysis according to claim 1 is characterized in that: In step 1, the second time period of data resampling is much shorter than the first time period.

5. The pressure early warning method based on threshold value and state trend analysis according to claim 1 is characterized in that: In step 2, in the moving average method of the sliding window, the relationship between the durations of the various time periods is that the third time period is shorter than the fourth time period, the fourth time period is shorter than the fifth time period; and the second time period is much shorter than the third time period.

6. The stress early warning method based on threshold value and state trend analysis according to claim 1 is characterized in that: In step 2, the rule for identifying an upward trend or a downward trend is configured as follows: if the current mean is greater than the previous mean, it is marked as an upward trend; if the current mean is less than the previous mean, it is marked as a downward trend.

7. The stress early warning method based on threshold value and state trend analysis according to claim 1 is characterized in that: In step 2, the functional relationship between the trend target value and the trend duration is constructed as follows: ； in x is the current real-time pressure value, y The trend target value set in advance. k is the pressure change rate of the two times before the current moment, is the pressure value 2 unit times before the current moment, and the unit time is the second time period; is the specific value of the duration or trend duration corresponding to x at the current moment, is the desired variable value.

8. The stress early warning method based on threshold and state trend analysis according to claim 1 is characterized in that: In step 3, the relationship between the real-time value and the threshold value formed by setting the threshold interval in segments is configured as follows: Set the threshold interval [attention value lower limit, attention value upper limit, alarm value lower limit]. The thresholds in the threshold interval are arranged from small to large. When the real-time value is less than the attention value lower limit, the early warning mechanism is not triggered. When the attention value lower limit is less than the real-time value and less than the attention value upper limit, the early warning mechanism is triggered. When the real-time value is greater than the alarm value lower limit, the alarm information is pushed immediately.

9. A pressure early warning system based on threshold and state trend analysis, characterized in that: include: A data acquisition module is used to obtain pressure time series data from a real-time database; A data processing module, used for performing outlier processing and data resampling on the acquired data; the data processing module includes a moving average calculation function of a sliding window; The trend identification module identifies the upward or downward trend based on the calculated mean, counts the duration of the trend, and constructs a functional relationship between the trend target value and the duration; Based on the functional relationship between the target value and the duration, the time required to reach the target value at the current trend rate is calculated; The early warning judgment module sets the threshold intervals in segments according to the actual business scenario rules, and determines whether to trigger the early warning mechanism based on the relationship between the real-time value and the threshold.

10. A computer-readable storage medium, characterized in that: A computer program is stored, and when the computer program is executed, it is used to implement the pressure warning method based on threshold value and state trend analysis as described in any one of claims 1 to 8.

Images