Thermal power generating unit equipment operation parameter early warning method and device based on dynamic score and medium
Through a dynamic scoring method, an estimation model of the operating parameters of the thermal power unit is established, dynamic scoring is calculated and early warning is provided, which solves the problem of difficulty in sensing the deviation of operating parameters from the normal state in the prior art, and improves the timeliness and accuracy of fault warnings.
Patent Information
- Application Number
- CN202411666716.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-21
- Publication Date
- 2025-02-25
AI Technical Summary
The prior art is difficult to sense in advance that the thermal power unit operating parameters deviate from the normal state, resulting in an alarm being triggered only after a fault occurs, and it is more passive to deal with abnormal working conditions.
The operation parameter warning method based on dynamic scoring is adopted, and by obtaining fixed high and low alarm limits, establishing an estimation model of operation parameters, calculating dynamic scores, and early warnings are made based on the scoring size.
An early warning of the operating parameters of the thermal power unit deviating from the normal state is realized, and the operation personnel are reminded to intervene in a timely manner, which improves the safety and reliability of the unit.
Smart Images
Figure CN119512001A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of intelligent early warning for coal-fired thermal power generation units, and in particular to a method and device for early warning of equipment operating parameters of thermal power generation units based on dynamic scoring. Background Art
[0002] In order to ensure the safe and economical operation of thermal power units, it is necessary to control the operating parameters of the units within the normal range. With the development of economy and technology, the scale of power generation units has gradually increased, and the operating parameters that need to be monitored have also increased dramatically. When monitoring massive operating parameters, it is difficult for operators to pay attention to the abnormal change trend of parameters in time. Often, the alarm signal can only be found after the fault occurs, and it is difficult to realize the early detection and intervention of abnormal conditions.
[0003] Traditional operating parameter early warning uses a fixed value alarm method, setting fixed alarm upper and lower limits for operating parameters according to the equipment and system operating safety boundaries. When the operating parameters exceed the upper / lower limits, the corresponding alarm, adjustment and protection logic are triggered. The development of generator set equipment or system failures is usually gradual, and the operating parameters usually gradually deviate from normal operating conditions, eventually reaching the alarm value and triggering the operation alarm, making it difficult for operators to observe the fluctuation range and degradation trend of equipment parameters in a timely manner. Parameter monitoring and alarm view each measuring point in isolation, and the alarm upper and lower limits of the parameters are usually fixed limits, resulting in the inability to perceive the deviation of the equipment or system from the normal operating state in advance. It is usually discovered after the failure occurs and the operating parameter exceeds the limit and triggers the alarm signal, and the abnormal condition handling is relatively passive.
[0004] The existing technology studies the causes and mechanism models of failures in thermal power plant operation equipment to achieve early warning of abnormal conditions. However, there are many causes and related parameters that lead to failures, and different causes and parameters have different degrees of influence on the occurrence of failures. In addition, the process flow of thermal power plants is complex and there are a lot of interferences. The mechanism of failure is often difficult to accurately describe, which leads to good theoretical effects of this method, but it is difficult to apply it in practice and difficult to promote it on a large scale. A more general parameter early warning method that can adapt to actual industrial production processes is needed. Summary of the invention
[0005] The purpose of the present invention is to propose an operating parameter early warning method, device and medium based on dynamic scoring to solve the problems of using fixed limit value alarm and being unable to perceive the deviation of operating parameters from normal operating conditions during the monitoring of operating parameters of thermal power units.
[0006] The purpose of the present invention can be achieved by the following technical solutions:
[0007] As a first aspect of the present invention, a method for early warning of equipment operating parameters of a thermal power unit based on dynamic scoring is provided, the steps comprising:
[0008] Obtain the fixed high and low alarm limits of the unit's operating parameters as the static high and low limits for calculating the dynamic score;
[0009] Read historical operation data, obtain relevant variables of operation parameters through correlation analysis, and establish a first estimation model of relevant variables for operation parameters;
[0010] The first estimation model is used to calculate a first estimation value of the operating parameter at each moment, and based on the deviation between the first estimation value and the actual value of the operating parameter, abnormal data points are determined and removed;
[0011] A second estimation model of the operating parameters for the relevant variables is established with the remaining data after removing the abnormal data points, and a second estimated value of the operating parameter is calculated; the maximum value of the deviation between the second estimated value of the operating parameter and the actual value is taken as the dynamic upper limit for calculating the dynamic score, and the minimum value of the deviation is taken as the dynamic lower limit;
[0012] Read the real-time values of relevant variables and operating parameters, and calculate the real-time estimated values of the operating parameters through the second estimation model; based on the actual values of the operating parameters and the deviation between the real-time estimated values and the actual values, calculate the dynamic score of the parameters according to the static upper limit, static lower limit, dynamic upper limit and dynamic lower limit, and issue an early warning based on the size of the dynamic score.
[0013] As a preferred technical feature, the fixed high and low alarm limits of the operating parameters are obtained based on the generator set alarm and protection setting list.
[0014] As a preferred technical feature, the method for establishing the estimation model of the relevant variables for the operating parameters includes linear or nonlinear fitting methods such as least squares method, random forest regression and neural network.
[0015] As a preferred technical feature, the process of determining and removing abnormal data points is as follows:
[0016] Calculate the deviation value e(k)k=1,2,…,N between the first estimated value and the actual value of the operating parameter at all times, where N is the total number of historical data, and sort the deviation values from small to large according to the absolute value;
[0017] Set the outlier judgment probability P a , sorted by absolute value and placed in P a The corresponding data points after N are determined as abnormal data points and removed.
[0018] As a preferred technical feature, the calculation rules of the dynamic score are as follows:
[0019] When the deviation between the estimated value and the actual value of the operating parameter is between the dynamic limits, the operating parameter is judged to be in a normal state and the score is 100;
[0020] When the deviation between the estimated value and the actual value of the operating parameter exceeds the dynamic limit, but the real-time value of the operating parameter does not exceed the static limit, it is determined that the operating parameter has an early abnormal sign, and the score gradually decreases;
[0021] When the operating parameter exceeds the static limit, the operating parameter is judged to be in an abnormal state and the score is 0.
[0022] As a preferred technical feature, when the deviation between the estimated value and the actual value of the operating parameter exceeds the dynamic limit, but the real-time value of the operating parameter does not exceed the static limit, the scoring rule is:
[0023]
[0024] Where e(0) is the estimated value of the operating parameter Deviation from the actual value y(0); H sta , L sta are the static upper limit and the static lower limit respectively; H dyn , L dyn They are dynamic high limit and dynamic low limit respectively.
[0025] As a preferred technical feature, the rules for issuing warnings based on the size of the dynamic score are as follows:
[0026] Set the upper and lower alarm limits; when the score is greater than the upper limit, the device is judged to be in normal condition; when the score is between the upper and lower limits, the device is judged to have deviated from normal operating conditions; when the score is less than the lower limit, the device is judged to be close to or has failed.
[0027] As a preferred technical feature, the method displays the device in different colors in the monitoring screen based on the different states of the device.
[0028] As a second aspect of the present invention, a thermal power unit equipment operating parameter early warning device based on dynamic scoring is provided, comprising a memory, a processor, and a program stored in the memory, characterized in that when the processor executes the program, the thermal power unit equipment operating parameter early warning method based on dynamic scoring as described above is implemented.
[0029] As a third aspect of the present invention, a storage medium is provided, on which a program is stored, characterized in that when the program is executed, the above-mentioned thermal power unit equipment operating parameter early warning method based on dynamic scoring is implemented.
[0030] Compared with the prior art, the present invention has the following beneficial effects:
[0031] The present invention is based on the massive historical operating data of the generator set, according to the relevant coupling relationship between the operating parameters, using statistical algorithms and big data processing methods to evaluate the health status scores of the operating parameters in real time. When the parameters deviate from the normal operating conditions but have not reached the traditional alarm upper and lower limits, the dynamic scoring is used to determine whether the operating parameters are normal. This can achieve early warning of abnormal conditions, remind operating personnel to intervene in time, and improve the safety and reliability of the unit. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] Figure 1 It is a structural schematic diagram of the present invention;
[0033] Figure 2 This is a schematic diagram of dynamic scoring calculation of the present invention;
[0034] Figure 3 The actual value and warning limit curve of the flue gas temperature at the outlet of air preheater A in one embodiment of the present invention;
[0035] Figure 4 It is a dynamic scoring curve of the flue gas temperature at the outlet of air preheater A in one embodiment of the present invention. DETAILED DESCRIPTION
[0036] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.
[0037] Example 1
[0038] The present invention proposes a thermal power unit equipment operating parameter early warning method based on dynamic scoring for thermal power unit operating parameter monitoring, which is aimed at the problem that it is difficult to monitor the key operating parameters of the equipment and it is impossible to perceive abnormal working conditions in advance. The data analysis modeling method is adopted to extract the main relevant variables of the thermal power unit equipment operating parameters, and an operating parameter estimation model is established based on massive historical operating data. A model-based operating parameter dynamic scoring method is designed to accurately reflect the unit operation status through a scoring early warning system. Figure 1 As shown, the specific technical methods are as follows:
[0039] S1. Based on the generator set alarm and protection setting list, obtain the fixed high and low alarm limits of the operating parameter y as the static high limit H required for calculating the dynamic score sta and static lower limit L sta .
[0040] S2, read the historical operation data, and select n main related variables x1, x2, ..., x of the operation parameter y through correlation analysis. nBased on historical operation data, x1, x2,…, x are established through linear or nonlinear fitting methods such as least squares method, random forest regression and neural network. n Estimation model for the running parameters y.
[0041] S3, the relevant parameter values at time k in the historical operation data are expressed as x1(k), x2(k),…, x n (k), calculate the estimated value of the operating parameter y at time k based on the estimation model in step 2) And compare it with the actual value y(k) of the operating parameter y, calculate the deviation value e(k) between the two, calculate all the deviation values at all times e(k)k=1,2,…,N, take the absolute value and sort |e(k)| from small to large, N is the total number of historical data.
[0042] S4. Set the outlier determination probability P a , sort |e(k)|k=1,2,…,N in The corresponding data points after are determined as abnormal data points. After removing the abnormal data points, the estimation model of the operating parameter y is re-established with the remaining N′ groups of data, and all deviation values e′(k) k=1,2,…,N′ are recalculated, and the maximum value e′ is taken. max =max[e′(k)]k=1,2,…,N′, as the dynamic upper limit H required for calculating dynamic scoring dyn =e max , minimum value e min =min[e′(k)]k=1,2,…,N′, dynamic lower limit L dyn =e min .
[0043] S5. Design the dynamic scoring calculation rules for the operating parameter y. Read the real-time values of the relevant parameters in real time (k=0 time) and calculate the estimated values of the operating parameters. The deviation e(0) from the actual value y(0) is used to calculate the dynamic score of the parameter according to the static upper limit, static lower limit, dynamic upper limit and dynamic lower limit. When the operating parameters are completely normal, the score S=100. The rules are as follows:
[0044] S=100,L dyn <e(0)<H dyn
[0045] When the deviation exceeds the dynamic limit, but the real-time value of the operating parameter does not exceed the static limit, the score gradually decreases. The rules are:
[0046]
[0047] When the operating parameters have exceeded the static limit, the score is 0, and the rules are:
[0048] S=0,y(0)≥H sta or y(0)≤L sta
[0049] The schematic diagram of dynamic score calculation is as follows Figure 2 shown.
[0050] S6. Set warning rules and perform warning operations according to the score. Set alarm limits L1 and L2. When the score S>L1, the equipment is judged to be in normal state, and the alarm display color in the monitoring screen is green. When the score L2≤S≤L1, the equipment is judged to deviate from the normal operating condition, and the alarm display color in the monitoring screen is yellow. The score S<L2 indicates that the equipment is close to or has failed, and the equipment color in the monitoring screen is red.
[0051] Example 2
[0052] As one of the specific implementation examples of the present invention, in this embodiment, the parameter warning process of the flue gas temperature at the outlet of air preheater A of a 1000MW ultra-supercritical coal-fired power unit is as follows:
[0053] S1. Based on the alarm and protection setting list of the generator set, obtain the fixed high and low alarm limits of the flue gas temperature y at the outlet of air preheater A, where the high protection value y max =31.7, as the static upper limit H required for calculating dynamic scores sta , low protection value y min =14.7, as the static lower limit L required to calculate the dynamic score sta .
[0054] S2, the data collection cycle is 10 seconds, read the historical data, and obtain the variables with obvious correlation with the outlet flue gas temperature y of air preheater A through correlation analysis, see Table 1:
[0055] Table 1 Variables associated with flue gas temperature at the outlet of air preheater A
[0056]
[0057]
[0058] By using the least squares method, we can initially establish x1, x2,…, x 10 Estimation model for y
[0059] S3. Calculate the difference between the estimated value and the actual value of the flue gas temperature at the outlet of air preheater A at each moment, e(k)k=1,2,…,N, take the absolute value of the difference and sort it from small to large.
[0060] S4. Set the outlier determination probability P a= 95%, |e(k)|k = 1, 2, ..., N is ranked The data is judged as abnormal data points. After removing the abnormal data points, the estimation model of the operating parameter y is re-established with the remaining N' groups of data. Recalculate the deviation value and take the maximum value e max =0.5989, and the dynamic upper limit H required for calculating the dynamic score is obtained. dyn =0.5989, minimum value e min =-0.6143, dynamic lower limit L dyn =-0.6143.
[0061] S5. Design the dynamic scoring calculation rules for operating parameter y. Read the real-time values of relevant parameters in real time and calculate the estimated value of the flue gas temperature at the outlet of air preheater A. The parameter dynamic score is calculated based on the actual value y(0), the static upper limit, the static lower limit, the dynamic upper limit, and the dynamic lower limit with the deviation e(0).
[0062] When the flue gas temperature at the outlet of air preheater A is judged to be completely normal, the score S=100, and the rule is:
[0063] S=100,L dyn <e(0)<H dyn
[0064] When the deviation exceeds the dynamic limit, but the real-time value of the flue gas temperature at the outlet of air preheater A does not exceed the static limit, the score gradually decreases, and the rules are:
[0065]
[0066] When the outlet flue gas temperature of air preheater A exceeds the static limit, the score is 0 and the rules are:
[0067] S=0,y(0)≥H sta or y(0)≤L sta
[0068] S6. Set the warning rules and perform warning operations according to the score. Set the alarm limit L1=80, L2=60. When the score S>L1, the device is judged to be in normal state, and the device color in the monitoring screen is green. When the score is L2≤S≤L1, the device is judged to deviate from the normal operating condition, and the device color in the monitoring screen is yellow. The score S<L2 indicates that the device is close to or has failed, and the device color in the monitoring screen is red.
[0069] The actual value and dynamic limit of the flue gas temperature at the outlet of air preheater A after running for a period of time. The static limit is shown in Figure 3 .
[0070] The corresponding dynamic scoring curve calculated according to the scoring rule in step (5) is as follows: Figure 4 shown.
[0071] Example 2
[0072] As a second aspect of the present invention, the present application also provides an electronic device, comprising: one or more processors; a memory for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the above-mentioned thermal power unit equipment operating parameter early warning method based on dynamic scoring. In addition to the above-mentioned processor, memory and interface, any device with data processing capability in which the device in the embodiment is located may also include other hardware according to the actual function of the device with data processing capability, which will not be described in detail.
[0073] Example 3
[0074] As a third aspect of the present invention, the present application also provides a computer-readable storage medium having computer instructions stored thereon, which, when executed by a processor, implements the above-mentioned thermal power unit equipment operating parameter early warning method based on dynamic scoring. The computer-readable storage medium may be an internal storage unit of any device with data processing capability described in any of the aforementioned embodiments, such as a hard disk or a memory. The computer-readable storage medium may also be an external storage device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), an SD card, a flash card (Flash Card), etc. equipped on the device. Furthermore, the computer-readable storage medium may also include both an internal storage unit and an external storage device of any device with data processing capability. The computer-readable storage medium is used to store the computer program and other programs and data required by any device with data processing capability, and may also be used to temporarily store data that has been output or is to be output.
[0075] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.
Claims
1. A method for early warning of equipment operating parameters of thermal power units based on dynamic scoring, characterized in that the steps include: Obtain the fixed high and low alarm limits of the unit operating parameters as the static high and low limits for calculating the dynamic score; Read historical operation data, obtain relevant variables of operation parameters through correlation analysis, and establish a first estimation model of relevant variables for operation parameters; The first estimation model is used to calculate a first estimation value of the operating parameter at each moment, and based on the deviation between the first estimation value and the actual value of the operating parameter, abnormal data points are determined and removed; Using the remaining data after removing the abnormal data points, a second estimation model of the operating parameters for the relevant variables is established, and a second estimated value of the operating parameters is calculated; The maximum value of the deviation between the second estimated value and the actual value of the operating parameter is taken as the dynamic upper limit for calculating the dynamic score, and the minimum value of the deviation is taken as the dynamic lower limit; Reading the real-time values of relevant variables and operating parameters, and calculating the real-time estimated values of the operating parameters through the second estimation model; Based on the actual value of the operating parameter and the deviation between the real-time estimated value and the actual value, the dynamic score of the parameter is calculated according to the static upper limit, static lower limit, dynamic upper limit and dynamic lower limit, and an early warning is issued based on the size of the dynamic score.
2. The method for early warning of thermal power plant equipment operating parameters based on dynamic scoring according to claim 1 is characterized in that: The fixed high and low alarm limits of the operating parameters are obtained based on the generator set alarm and protection setting list.
3. The method for early warning of thermal power plant equipment operating parameters based on dynamic scoring according to claim 1 is characterized in that: Methods for establishing the estimation model of the relevant variables to the operating parameters include linear or nonlinear fitting methods such as least squares method, random forest regression and neural network.
4. The method for early warning of thermal power plant equipment operating parameters based on dynamic scoring according to claim 1 is characterized in that: The specific process of determining and removing abnormal data points is as follows: Calculate the deviation value e(k)k=1,2,…,N between the first estimated value and the actual value of the operating parameter at all times, where N is the total number of historical data, and sort the deviation values from small to large according to the absolute value; Set the outlier judgment probability P a , sorted by absolute value and placed in P a The corresponding data points after N are determined as abnormal data points and removed.
5. The method for early warning of thermal power plant equipment operating parameters based on dynamic scoring according to claim 1 is characterized in that: The calculation rules of the dynamic score are as follows: When the deviation between the estimated value and the actual value of the operating parameter is between the dynamic limits, the operating parameter is judged to be in a normal state; When the deviation between the estimated value and the actual value of the operating parameter exceeds the dynamic limit, but the real-time value of the operating parameter does not exceed the static limit, it is determined that the operating parameter has an early abnormal sign, and the score gradually decreases; When the operating parameter exceeds the static limit, the operating parameter is judged to be in an abnormal state and the score is 0.
6. The method for early warning of thermal power plant equipment operating parameters based on dynamic scoring according to claim 5 is characterized in that: When the deviation between the estimated value and the actual value of the operating parameter exceeds the dynamic limit, but the real-time value of the operating parameter does not exceed the static limit, the scoring rule is: Where e(0) is the estimated value of the operating parameter Deviation from the actual value y(0); H sta , L sta They are the static upper limit and the static lower limit respectively; H dyn , L dyn They are dynamic high limit and dynamic low limit respectively.
7. The method for early warning of thermal power plant equipment operating parameters based on dynamic scoring according to claim 1 is characterized in that: The specific rules for issuing warnings based on the size of the dynamic score are as follows: Set the upper and lower alarm limits; when the score is greater than the upper limit, the device is judged to be in normal condition; when the score is between the upper and lower limits, the device is judged to have deviated from normal operating conditions; when the score is less than the lower limit, the device is judged to be close to or has failed.
8. The method for early warning of thermal power plant equipment operating parameters based on dynamic scoring according to claim 7 is characterized in that: The method displays the device in different colors in the monitoring screen based on different states of the device.
9. A thermal power unit equipment operation parameter early warning device based on dynamic scoring, comprising a memory, a processor, and a program stored in the memory, characterized in that: When the processor executes the program, the method for early warning of operating parameters of thermal power unit equipment based on dynamic scoring as described in any one of claims 1 to 8 is implemented.
10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for early warning of operating parameters of thermal power unit equipment based on dynamic scoring as described in any one of claims 1 to 8 is implemented.