A unit equipment parameter early warning method, device and storage medium

Abstract

The present invention relates to the field of information technology, and specifically to a method, device and storage medium for warning parameters of a unit equipment, comprising the following steps: reading historical data of unit equipment parameters; receiving warning items added by users; obtaining the alarm limit of each target parameter of each warning item under corresponding working conditions according to historical data; establishing an equipment parameter prediction model, and using historical data to train the parameter prediction model; receiving the warning cycle set by the user; after the warning cycle arrives, obtaining the predicted values ​​of all parameters of the unit equipment according to the equipment parameter prediction model; reading each warning item in turn, judging whether the predicted value of the target parameter exceeds the alarm limit, if so, issuing an alarm, otherwise, entering the next warning item. The beneficial technical effects of the present invention include: dynamically calculating the upper and lower alarm limits of the target indicator, adapting to the warning under different working conditions, and effectively improving the accuracy of the warning.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a method, a device and a storage medium for warning parameters of a unit equipment. Background Art

[0002] The early warning of power plant unit equipment parameters can identify possible problems with equipment parameters in advance and ensure the normal operation of power plant unit equipment. Therefore, there is a practical demand for early warning of unit equipment parameters. At present, the early warning of power plant unit equipment parameters adopts the setting of unified high and low limits of indicator alarm. Because the unit has different high and low limits of equipment parameter alarms under different working conditions. However, the current technology cannot set the corresponding high and low limits according to different working conditions. This leads to false alarms or missed alarms, and it is difficult to ensure the accuracy and reliability of equipment parameter early warning. At present, there is a lack of real-time value prediction schemes for power plant unit equipment parameters, and it is impossible to combine the predicted values ​​to judge in advance whether the unit equipment may have faults. It can only be processed after the unit equipment issues a fault alarm, which reduces the operating efficiency and reliability of the unit and affects the stable supply of electricity. In view of the above situation, it is necessary to study new early warning technologies for power plant unit equipment parameters.

[0003] After searching, the applicant found that CN114637791A is the prior art closest to the present application, and its publication date is June 17, 2022. It records a power plant equipment early warning method based on similarity. This method calculates similarity based on the current important parameters of the power plant equipment and the historical normal state database, discovers equipment abnormalities in advance, and reduces the failure rate of power plant equipment. Specifically, it includes the following steps: Step 1: Collect historical data, collect important feature data of the early warning equipment, select stable operating condition data, and obtain a stable operating condition data set; Step 2: Based on the stable operating condition data set of step 1, clean the abnormal data; Step 3: Based on the cleaned data obtained in step 2, select data within the similar normal range according to the key feature data of the unit; Step 4: Determine whether the current operating state is a stable operating condition. If it is a stable operating condition, calculate the similarity value between the current feature value of the early warning equipment and each piece of data in step 3, select the maximum value of the similarity value, and start the early warning if the value is less than the set threshold. Its technical solution uses a similarity calculation method to take into account the coupling relationship between parameters, and measures the health of the equipment by calculating the similarity of distance information and direction information, thereby providing fault warning for power plant equipment. However, its technical solution cannot set different warning values ​​for different unit operating conditions, and still cannot avoid false alarms or missed alarms when the unit is operating under different operating conditions. Summary of the invention

[0004] The technical problem to be solved by the present invention is that there is currently a lack of a solution for dynamically setting warning values ​​that is suitable for different operating conditions of the unit. A method, device and storage medium for warning of unit equipment parameters are proposed, which can dynamically set alarm limits under different unit operating conditions to improve the accuracy and reliability of equipment parameter warnings.

[0005] To solve the above technical problems, the present invention adopts the following technical solution: a unit equipment parameter early warning method, comprising the following steps:

[0006] Read historical data of unit equipment parameters;

[0007] Receive a warning item added by a user, wherein the warning item includes an operating condition and a plurality of target parameters;

[0008] Obtain the alarm limit value of each target parameter for each warning item under the corresponding working conditions based on historical data;

[0009] Establishing an equipment parameter prediction model and using historical data to train the parameter prediction model;

[0010] Receive the warning cycle set by the user;

[0011] After the early warning period arrives, the predicted values ​​of all parameters of the unit equipment are obtained according to the equipment parameter prediction model;

[0012] Read each warning item in turn. If the operating condition of the unit equipment meets the operating condition of the warning item, further determine whether the predicted value of the target parameter exceeds the alarm limit. If so, issue an alarm. Otherwise, if the operating condition of the unit equipment does not meet the operating condition of the warning item or the predicted value of the target parameter does not exceed the alarm limit, proceed to the next warning item.

[0013] Preferably, the method for calculating the alarm limit of the target parameter comprises:

[0014] Read all data that meets the working conditions of the warning items from the historical data, and obtain all historical data of the corresponding target parameters;

[0015] Calculate the mean of the historical data for each target parameter and standard deviation s i ;

[0016] The upper limit of the target parameter alarm is The lower alarm limit of the target parameter is

[0017] Preferably, the value of k belongs to the interval [2,5].

[0018] Preferably, the value of k is 4.

[0019] Preferably, after reading the historical data of the unit equipment parameters, the historical data is preprocessed to remove outliers, and the missing values ​​are supplemented by interpolation of the values ​​before and after the missing values.

[0020] Preferably, the method for removing outliers includes:

[0021] Sort the data values ​​of historical data by generation time;

[0022] Set the window length to L1, read L1 / 2 data values ​​before and after the data value, and calculate the average value;

[0023] If the data value exceeds the interval [k1*mean value, k2*mean value], the data value is determined to be an outlier and is removed.

[0024] Preferably, the method for supplementing missing values ​​includes:

[0025] Sort the data values ​​of historical data by generation time;

[0026] Set the window length to L2, read L2 / 2 data values ​​before and after the missing value, and obtain L2 data values;

[0027] Interpolate missing values ​​using quadratic interpolation.

[0028] Preferably, the method for establishing the equipment parameter prediction model comprises:

[0029] Read the historical data of the unit equipment parameters and establish the training matrix K, K = [X1M (1) ,X2M (2) ,X3M (3) ,…,X (m) M (m) ], m represents the number of device parameters contained in the training matrix;

[0030] Establish the health matrix D, n represents the data row;

[0031] According to the training matrix K and the health matrix D, we can get M1, M2, M3, ..., M m The value of

[0032] Set the normalization factor according to the device parameters to be predicted;

[0033] According to the algorithm, the real-time value Y of the normalization factor is predicted, Y = [X 1(n+1) M (1) ,X 2(n+1) M (2) ,X 3(n+1) M (3) ,…,X (m)(n+1) M (m) ];

[0034] According to the real-time value of the predicted normalization factor Y and M1 to M m The value of X 1(n+1) ,X 2(n+1) ,X 3(n+1) ,…,X (m)(n+1) The value of is the value to be predicted of the equipment parameter.

[0035] A unit equipment parameter early warning device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the computer program implements a unit equipment parameter early warning method as described above when executed by the processor.

[0036] A storage medium stores a computer program, and when the computer program is executed by a processor, the computer program implements the above-mentioned unit equipment parameter early warning method.

[0037] The beneficial technical effects of the present invention include: the present invention collects online data of equipment parameters according to different set warning items and combines the current working conditions of the unit, dynamically calculates the upper and lower limits of target indicator alarms, and adapts to warnings under different working conditions. At the same time, it provides an equipment parameter prediction model, and issues warnings based on the predicted values ​​of equipment parameters, so that equipment parameter warnings have more advance time, effectively improving equipment working efficiency and the accuracy of equipment warnings under different working conditions.

[0038] Other features and advantages of the present invention will be disclosed in detail in the following specific embodiments and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The present invention will be further described below in conjunction with the accompanying drawings:

[0040] Figure 1 The figure is a flow chart of a method for warning parameters of a unit equipment according to an embodiment of the present invention.

[0041] Figure 2 The figure is a flow chart of a method for calculating alarm limits of target parameters according to an embodiment of the present invention.

[0042] Figure 3 The figure is a flow chart of a method for removing outliers according to an embodiment of the present invention.

[0043] Figure 4 The figure is a flow chart of a method for supplementing missing values ​​according to an embodiment of the present invention. DETAILED DESCRIPTION

[0044] The technical solutions of the embodiments of the present invention are explained and described below in conjunction with the drawings of the embodiments of the present invention, but the following embodiments are only preferred embodiments of the present invention, not all. Based on the embodiments in the implementation mode, other embodiments obtained by those skilled in the art without creative work are all within the protection scope of the present invention.

[0045] In the following description, terms such as "inside", "outside", "up", "down", "left", "right", etc. that indicate directions or positional relationships are only used to facilitate the description of the embodiments and simplify the description, and do not indicate or imply that the referred device or element must have a specific direction, be constructed and operate in a specific direction. Therefore, it should not be understood as a limitation of the present invention.

[0046] A unit equipment parameter early warning method, please refer to the attached Figure 1 , including the following steps:

[0047] Step A01) reading historical data of unit equipment parameters;

[0048] Step A02) receiving warning items added by the user, the warning items including working conditions and several target parameters;

[0049] Step A03) obtaining the alarm limit value of each target parameter of each warning item under corresponding working conditions according to historical data;

[0050] Step A04) Establishing a device parameter prediction model, and using historical data to train the parameter prediction model;

[0051] Step A05) receiving the warning cycle set by the user;

[0052] Step A06) After the early warning period arrives, the predicted values ​​of all parameters of the unit equipment are obtained according to the equipment parameter prediction model;

[0053] Step A07) reads each warning item in turn. If the operating condition of the unit equipment meets the operating condition of the warning item, it is further determined whether the predicted value of the target parameter exceeds the alarm limit. If it exceeds, an alarm is issued. Otherwise, if the operating condition of the unit equipment does not meet the operating condition of the warning item or the predicted value of the target parameter does not exceed the alarm limit, enter the next warning item.

[0054] Please refer to the attached Figure 2 , the method for calculating the alarm limit of the target parameter includes:

[0055] Step B01) reading all data of the working conditions that meet the warning items from the historical data to obtain all historical data of the corresponding target parameters;

[0056] Step B02) Calculate the mean of the historical data of each target parameter and standard deviation si ;

[0057] Step B03) The alarm upper limit value of the target parameter is The lower alarm limit of the target parameter is

[0058] Preferably, the value of k is in the interval [2,5]. Preferably, the value of k is selected as 4. That is, the upper limit of the alarm of the target parameter is The lower alarm limit of the target parameter is

[0059] Step A01) After reading the historical data of the unit equipment parameters, the historical data is preprocessed to remove outliers and to supplement the missing values ​​by interpolating the values ​​before and after the missing values.

[0060] Please refer to the attached Figure 3 , the methods for removing outliers include:

[0061] Step C01) sorting the data values ​​of the historical data according to the generation time;

[0062] Step C02) setting the window length L1, reading L1 / 2 data values ​​before and after the data value respectively, and calculating the average value;

[0063] Step C03) If the data value exceeds the interval [k1*average value, k2*average value], the data value is determined to be an outlier and is removed.

[0064] Please refer to the attached Figure 4 , the methods for filling missing values ​​include:

[0065] Step D01) sorting the data values ​​of the historical data according to the generation time;

[0066] Step D02) Setting the window length L2, reading L2 / 2 data values ​​before and after the missing value, respectively, to obtain L2 data values;

[0067] Step D03) interpolates missing values ​​using quadratic interpolation.

[0068] The working condition tag0 meets: K1≤tag0≤K2, K1 and K2 are the set conditions. Filter all data of the target parameters that meet the working condition. The target parameters are tag1 to tagm, tag1=[x 11 ,x 12 ,x 13 ,..,x 1n ],tag1=[x 21 ,x 22 ,x 23 ,..,x 2n ],…,tagm=[x m1 ,xm2 ,x m3 ,..,x mn ], calculate the mean and standard deviation of the target parameter data that meets the conditions: in Represents the mean.

[0069] Calculate the standard deviation:

[0070] The dynamic alarm upper limit is:

[0071] The dynamic alarm lower limit is:

[0072] For example:

[0073] Assuming that the historical data of the selected target parameters under the corresponding working conditions is [100, 90, 110], the average value is: Find the standard deviation: s i =8.16. Then the upper limit of dynamic alarm is: 100+8.16*4=132, and the lower limit of dynamic alarm is: 100-8.16*4=67.34.

[0074] The alarm limit of the target parameter is dynamically determined based on the historical data of the normal operation of the target parameter under the corresponding working conditions, and the alarm limit of each target parameter under different working conditions is obtained, which has higher accuracy and adaptability. It helps to reduce missed alarms and false alarms and improve the accuracy and reliability of the unit equipment parameter early warning.

[0075] Methods for establishing equipment parameter prediction models include:

[0076] Read the historical data of the unit equipment parameters and establish the training matrix K, K = [X1M (1) ,X2M (2) ,X3M (3) ,…,X (m) M (m) ], m represents the number of device parameters contained in the training matrix;

[0077] Establish the health matrix D, n represents the data row;

[0078] According to the training matrix K and the health matrix D, we can get M1, M2, M3, ..., M m The value of

[0079] Set the normalization factor according to the device parameters to be predicted;

[0080] According to the algorithm, the real-time value Y of the normalization factor is predicted, Y = [X 1(n+1) M (1) ,X2(n+1) M (2) ,X 3(n+1) M (3) ,…,X (m)(n+1) M (m) ];

[0081] According to the real-time value of the predicted normalization factor Y and M1 to M m The value of X 1(n+1) ,X 2(n+1) ,X 3(n+1) ,…,X (m)(n+1) The value of is the value to be predicted of the equipment parameter.

[0082] A unit equipment parameter early warning device, the unit equipment parameter early warning device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, and when the computer program is executed by the processor, a unit equipment parameter early warning method as described above is implemented.

[0083] A storage medium stores a computer program, and when the computer program is executed by a processor, the aforementioned unit equipment parameter early warning method is implemented.

[0084] The beneficial technical effects of this embodiment include: the present invention collects online data of equipment parameters according to different set warning items and combines the current working conditions of the unit, dynamically calculates the upper and lower limits of target indicator alarms, and adapts to warnings under different working conditions. At the same time, it provides an equipment parameter prediction model, and issues warnings based on the predicted values ​​of equipment parameters, so that equipment parameter warnings have more advance time, effectively improving equipment work efficiency and the accuracy of equipment warnings under different working conditions.

[0085] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Those skilled in the art should understand that the present invention includes but is not limited to the contents described in the drawings and the above specific embodiments. Any modification that does not deviate from the functional and structural principles of the present invention will be included in the scope of the claims.

Claims

1. A unit equipment parameter early warning method, characterized in that: The following steps are involved: Read historical data of unit equipment parameters; Receive a warning item added by a user, wherein the warning item includes an operating condition and a plurality of target parameters; Obtain the alarm limit value of each target parameter for each warning item under the corresponding working conditions based on historical data; Establishing an equipment parameter prediction model, and using historical data to train the parameter prediction model; Receive the warning cycle set by the user; After the early warning period arrives, the predicted values ​​of all parameters of the unit equipment are obtained according to the equipment parameter prediction model; Read each warning item in turn. If the operating condition of the unit equipment meets the operating condition of the warning item, further determine whether the predicted value of the target parameter exceeds the alarm limit. If so, issue an alarm. Otherwise, if the operating condition of the unit equipment does not meet the operating condition of the warning item or the predicted value of the target parameter does not exceed the alarm limit, proceed to the next warning item. Methods for establishing equipment parameter prediction models include: Read the historical data of the unit equipment parameters and establish the training matrix K, K = [X1M (1) ,X2M (2) ,X3M (3) ,…,X (m) M (m) ], m represents the number of device parameters contained in the training matrix; Establish the health matrix D, n represents the data row; According to the training matrix K and the health matrix D, we can get M1, M2, M3, ..., M m The value of Set the normalization factor according to the device parameters to be predicted; According to the algorithm, the real-time value Y of the normalization factor is predicted, Y = [X 1(n+1) M (1) ,X 2(n+1) M (2) ,X 3(n+1) M (3) ,…,X (m)(n+1) M (m) ]; According to the real-time value of the predicted normalization factor Y and M1 to M m The value of X 1(n+1) ,X 2(n+1) ,X 3(n+1) ,…,X (m)(n+1) The value of is the value to be predicted of the equipment parameter.

2. A unit equipment parameter early warning method according to claim 1, characterized in that: Methods for calculating alarm limits for target parameters include: Read all the data that meets the working conditions of the warning items from the historical data, obtain all the historical data of the corresponding target parameters; calculate the mean of the historical data of each target parameter and standard deviation s i ; The upper limit of the target parameter alarm is The lower alarm limit of the target parameter is 3. A unit equipment parameter early warning method according to claim 2, characterized in that: The value of k belongs to the interval [2,5].

4. A unit equipment parameter early warning method according to claim 2, characterized in that: The value of k is 4.

5. A unit equipment parameter early warning method according to any one of claims 1 to 4, characterized in that: After reading the historical data of the unit equipment parameters, the historical data is preprocessed to remove outliers, and the missing values ​​are supplemented by interpolating the values ​​before and after the missing values.

6. A unit equipment parameter early warning method according to claim 5, characterized in that: Methods for removing outliers include: Sort the data values ​​of historical data by generation time; Set the window length to L1, read L1 / 2 data values ​​before and after the data value, and calculate the average value; If the data value exceeds the interval [k1*mean value, k2*mean value], the data value is determined to be an outlier and is removed.

7. A unit equipment parameter early warning method according to claim 5, characterized in that: Methods for filling missing values ​​include: Sort the data values ​​of historical data by generation time; Set the window length to L2, read L2 / 2 data values ​​before and after the missing value, and obtain L2 data values; Interpolate missing values ​​using quadratic interpolation.

8. A unit equipment parameter early warning device, characterized in that: The unit equipment parameter warning device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, a unit equipment parameter warning method as described in any one of claims 1 to 7 is implemented.

9. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method for early warning of unit equipment parameters as described in any one of claims 1 to 7 is implemented.

Images