Intelligent diagnosis method based on cold end offline data
Through intelligent diagnostic methods based on offline data of the cold end, the VAR model is used to predict the cold end operating conditions and optimize the cold end operating conditions, which solves the problems of monitoring lag and false alarms and missed reports in the existing cold end system, and achieves more efficient and reliable cold end system operation and maintenance management.
Patent Information
- Application Number
- CN202411912792.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-24
- Publication Date
- 2025-04-15
AI Technical Summary
The monitoring of existing cold-end systems relies on regular manual inspections or fixed data collection points, making it difficult to capture instantaneous abnormalities and sudden failures, resulting in lag in control and optimization, and it is difficult to accurately identify potential problems, resulting in frequent missed or false positives.
Intelligent diagnostic methods based on offline data of the cold end are adopted, and key data such as circulating water flow and strengthening the speed of the heat exchange device are transmitted to the server for processing. The intelligent diagnostic model based on VAR model is used to predict the cold end working conditions, and the model is optimized through verification and machine learning to form a mapping relationship to improve prediction capabilities and diagnostic accuracy.
It realizes predictive discovery of potential problems, reduces the lag reaction time before the failure occurs, improves the operating efficiency and reliability of the cold-end system, reduces operation and maintenance costs, and enhances the prediction ability and diagnostic accuracy of future working conditions.
Smart Images
Figure BDA0005206147190000031 
Figure BDA0005206147190000041 
Figure BDA0005206147190000042
Abstract
Description
Technical Field
[0001] The present invention relates to the field of cold end intelligent diagnosis, and in particular to an intelligent diagnosis method based on cold end offline data. Background Art
[0002] The cold end system is an important part of the generator set, including the low-pressure cylinder body and its heat recovery steam extraction system, exhaust port and condenser system, circulating water system, sealing system and vacuum system, cooling tower system and other subsystems. The core purpose of this system is to convert the exhaust gas of the steam turbine into condensed water, so that the power generation process can heat the water into steam again to achieve continuous power generation. The operating quality of the cold end system is directly related to the vacuum degree of the generator set, and the vacuum degree or back pressure of the steam turbine of a large generator set has a significant impact on its energy consumption level and economy. Research data shows that when the back pressure exceeds the critical back pressure, the coal consumption of power generation will increase by 2-3g / KWh for every 1KPa increase in back pressure. Therefore, the operating goal of the cold end system is to strive to achieve the best vacuum degree or the lowest back pressure under certain circulating water system power consumption conditions.
[0003] With the continuous development of industrial technology, cold end systems are increasingly used in various types of equipment, and the stability and reliability of their operating status are crucial to the performance of the entire system. Many existing cold ends rely on regular manual inspections or timed data collection, which makes it difficult to capture instantaneous anomalies and sudden failures during the operation of the cold end, resulting in a lag in control and optimization. In some cases, potential problems in the cold end system may not be accurately identified, resulting in frequent omissions or false alarms. Summary of the invention
[0004] The present invention aims to provide an intelligent diagnosis method based on cold-end offline data, and performs intelligent diagnosis by collecting and analyzing offline data.
[0005] To achieve the above object, the present invention adopts the following technical solution: comprising the following steps:
[0006] An intelligent diagnosis method based on cold end offline data comprises the following steps:
[0007] Acquire cold end offline data, wherein the cold end offline data includes offline flow rate of circulating water and speed data of enhanced heat exchange device;
[0008] Transmitting cold-end offline data, where the data transmission end sends the cold-end offline data to the server end; the server end adds an offline timestamp to the acquired cold-end offline data to obtain an offline data set;
[0009] To generate cold-end offline diagnosis results, the server inputs the offline data set into the intelligent diagnosis model and outputs the cold-end offline diagnosis results including the offline timestamp.
[0010] Preferably, as an improvement, the intelligent diagnosis model is based on a VAR model, and through an offline data set input, the intelligent diagnosis model performs trend prediction on the cold end working condition to obtain a prediction result.
[0011] Preferably, as an improvement, the server verifies the prediction result, and the verification is to compare the prediction result with the corresponding offline data set.
[0012] Preferably, as an improvement, a mapping relationship is formed between the prediction result and the offline data set of the cold end through verification on the server side, and the offline data set of the cold end is used for machine learning of the intelligent diagnosis model.
[0013] Preferably, as an improvement, the cold end offline diagnosis result output by the server includes a verification prediction result, which is a verification prediction result obtained by the server through a historical offline data set.
[0014] Preferably, as an improvement, the offline data set includes Y1: offline flow of circulating water at time point t; Y2: heat load of condenser; and Y3: efficiency of cooling tower.
[0015] Preferably, as an improvement, the prediction result at time t+1 is a prediction value calculated by using the offline data sets at time t and time t-1.
[0016] Preferably, as an improvement, the offline data set at time t+1 is used to remove abnormal points in the offline data sets at time t and time t-1.
[0017] Preferably, as an improvement, the offline data set at time t+1 is used to correct the abnormal points in the offline data sets at time t and time t-1.
[0018] The principle and advantages of this solution are: by acquiring the key data such as the circulating water flow rate of the cold end and the speed of the enhanced heat exchange device offline, the data are transmitted to the server for processing. This overcomes the lag and incompleteness of traditional cold end system monitoring, which often relies on regular manual inspections or fixed data collection points. By using the intelligent diagnosis model based on the VAR model to predict the trend of the cold end working conditions, potential problems can be foreseen, so that preventive measures can be taken before the problems occur. The server verifies the prediction results, and by comparing the prediction results with the actual offline data set, the accuracy of the model can be verified and continuously optimized. By mapping the offline data set of the cold end with the prediction results, and using these data for machine learning, the intelligent diagnosis model can be continuously optimized. This enables the model to gradually adapt to the changes in the cold end system and improve the predictive ability and diagnostic accuracy of future working conditions. The accuracy and reliability of the diagnostic model are ensured, and the model can be continuously adapted to the changes in the cold end system through machine learning optimization, which improves the predictive ability and diagnostic accuracy of future working conditions and better serves the operation and maintenance management of the cold end system. Using offline data sets at subsequent moments to remove or correct abnormal points in data at previous moments helps eliminate noise and errors in the data and improve data quality, which helps improve the operating efficiency of the cold-end system, extend equipment life, and reduce operation and maintenance costs.
[0019] In summary, this intelligent diagnosis method based on cold-end offline data has significantly improved the operating efficiency and reliability of the cold-end system through technical effects such as predictability, accuracy improvement, machine learning optimization, abnormal point processing, and comprehensive decision support, providing a strong guarantee for the company's safe production and energy conservation and emission reduction. DETAILED DESCRIPTION
[0020] The following is further described in detail through specific implementation methods:
[0021] The specific implementation steps are as follows:
[0022] An intelligent diagnosis method based on cold end offline data comprises the following steps:
[0023] Acquire cold end offline data, wherein the cold end offline data includes offline flow rate of circulating water and speed data of enhanced heat exchange device;
[0024] Transmitting cold-end offline data, where the data transmission end sends the cold-end offline data to the server end; the server end adds an offline timestamp to the acquired cold-end offline data to obtain an offline data set;
[0025] To generate cold-end offline diagnosis results, the server inputs the offline data set into the intelligent diagnosis model and outputs the cold-end offline diagnosis results including the offline timestamp.
[0026] Wherein, the intelligent diagnosis model is based on the VAR model, and through the offline data set input, the intelligent diagnosis model performs trend prediction on the cold end working condition to obtain the prediction result.
[0027] The server verifies the prediction result, and the verification is to compare the prediction result with the corresponding offline data set.
[0028] Through verification on the server side, the prediction results are mapped to the offline data set on the cold side, and the offline data set on the cold side is used for machine learning of the intelligent diagnosis model.
[0029] The cold end offline diagnosis result output by the server includes a verification prediction result, which is a prediction result obtained by the server through a historical offline data set.
[0030] For example, to obtain cold end offline data, the following cold end offline data is obtained at a certain moment:
[0031] Circulating water offline flow: 1500m 3 / h;
[0032] Speed data of enhanced heat exchange device: 3000rpm.
[0033] The data transmission end sends the cold end offline data to the server. After receiving the data, the server adds an offline timestamp to each piece of data (assuming it is 2024-12-12T16:20:00Z) to obtain the offline data set:
[0034]
[0035]
[0036] Generate cold end offline diagnosis results: The server inputs the above offline data set into the intelligent diagnosis model based on the VAR model. The model predicts the trend of the cold end working condition. Assume that the prediction result is that the offline flow rate of circulating water will rise to 1550m in the next 10 minutes. 3 / h, the speed of the enhanced heat exchanger will remain unchanged. The server outputs the prediction result together with the offline timestamp (assuming it is 2024-12-12T16:20:00Z) to form the cold end offline diagnosis result:
[0037]
[0038] Verify the prediction results: The server verifies the prediction results, that is, compares the prediction results with the corresponding offline data set. Assume that the server uses a historical offline data set (that is, previously collected and timestamped data) to verify the accuracy of the model. For example, the server may find that the prediction error rate of the model under similar working conditions in the past is within 5%, so it believes that the current prediction result is reliable.
[0039] Forming a mapping relationship and performing machine learning: Through verification on the server side, a mapping relationship is formed between the prediction results and the offline datasets on the cold side. The server side can use these offline datasets with timestamps for continuous machine learning of the intelligent diagnosis model to optimize the model's prediction performance. For example, the server side uses algorithms such as gradient descent to adjust model parameters to minimize prediction errors.
[0040] The offline data set includes Y1 at time point t: offline flow of circulating water; Y2: heat load of condenser; and Y3: efficiency of cooling tower.
[0041] The intelligent diagnosis model is:
[0042]
[0043] Among them, c1, c2, c3 are constant terms, ij, k are lag term coefficients, and it is the error term.
[0044] The prediction result at time t+1 is the prediction value calculated using the offline data sets at time t and time t-1.
[0045] The offline data set at time t+1 is used to remove abnormal points in the offline data sets at time t and time t-1.
[0046] The offline data set at time t+1 is used to correct the abnormal points in the offline data sets at time t and time t-1.
[0047] For example, there is the following time series data, which records the offline data sets at time points t, t-1, and t+1, including Y1: offline flow rate of circulating water (unit: m 3 / h), Y2: condenser heat load (unit: MW), Y3: cooling tower efficiency (unit: %).
[0048]
[0049] Outlier Detection
[0050] Determine the value of Y1 (circulating water offline flow) at time point t, 1500m 3 / h is an abnormally high value. Similarly, we can detect whether there are abnormal points in other variables and other time points.
[0051] Outlier removal
[0052] If you choose to remove this outlier, you can delete the Y1 value at time t from the data set, or fill in the missing value. Here, just delete the value directly.
[0053] The dataset after removing outliers may look like this:
[0054]
[0055] In practical applications, directly deleting data points may not be the best choice because it will reduce the sample size of the data set. A better approach may be outlier correction.
[0056] If we choose to correct this outlier, we can use the data at time t+1 to help us find a more reasonable value to replace it. We can assume that the value of Y1 changes smoothly between adjacent time points and use linear interpolation or other more complex interpolation methods to estimate the value of Y1 at time t.
[0057] The dataset after correcting the outliers using linear interpolation might look like this:
[0058]
[0059] Here, we use (1200m 3 / h+1300m 3 / h) / 2=1250m 3 / h as the corrected value of Y1 at time t, but in order to keep it close to the original outlier value (and still be a "corrected" value), it is rounded to 1350m 3 / h.
[0060] Through the above process, we can see how to use the offline data set at time t+1 to detect, remove and correct the outliers in the data sets at time t and time t-1. In practical applications, these steps need to be adjusted according to the characteristics of the data and the objectives of the analysis.
[0061] The principle and advantages of this solution are: by acquiring the key data such as the circulating water flow rate of the cold end and the speed of the enhanced heat exchange device offline, the data are transmitted to the server for processing. This overcomes the lag and incompleteness of traditional cold end system monitoring, which often relies on regular manual inspections or fixed data collection points. By using the intelligent diagnosis model based on the VAR model to predict the trend of the cold end working conditions, potential problems can be foreseen, so that preventive measures can be taken before the problems occur. The server verifies the prediction results, and by comparing the prediction results with the actual offline data set, the accuracy of the model can be verified and continuously optimized. By mapping the offline data set of the cold end with the prediction results, and using these data for machine learning, the intelligent diagnosis model can be continuously optimized. This enables the model to gradually adapt to the changes in the cold end system and improve the predictive ability and diagnostic accuracy of future working conditions. The accuracy and reliability of the diagnostic model are ensured, and the model can be continuously adapted to the changes in the cold end system through machine learning optimization, which improves the predictive ability and diagnostic accuracy of future working conditions and better serves the operation and maintenance management of the cold end system. Using offline data sets at subsequent moments to remove or correct abnormal points in data at previous moments helps eliminate noise and errors in the data and improve data quality, which helps improve the operating efficiency of the cold-end system, extend equipment life, and reduce operation and maintenance costs.
[0062] The above is only an embodiment of the present invention, and the common knowledge such as the known specific technical schemes and / or characteristics in the scheme is not described in detail here. It should be pointed out that for those skilled in the art, several variations and improvements can be made without departing from the technical scheme of the present invention. In the present invention, unless otherwise clearly specified and limited, the terms "install", "connect", "connect", "fix" and the like 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 directly connected, or indirectly connected through an intermediate medium, or it can be the internal connection of two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances. The scope of protection claimed in this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.
Claims
1. An intelligent diagnosis method based on cold end offline data, characterized in that: The following steps are involved: Acquire cold end offline data, wherein the cold end offline data includes offline flow rate of circulating water and speed data of enhanced heat exchange device; Transmitting cold-end offline data: The data transmission end periodically sends the cold-end offline data to the server end; the server end adds a timestamp to the acquired cold-end offline data to obtain an offline data set; To generate cold-end offline diagnosis results, the server inputs the offline data set into the intelligent diagnosis model and outputs the cold-end offline diagnosis results including the offline timestamp.
2. The intelligent diagnosis method based on cold end offline data according to claim 1, characterized in that: The intelligent diagnosis model is based on the VAR model. Through the input of the offline data set, the intelligent diagnosis model performs trend prediction on the cold end working condition to obtain the prediction result.
3. The intelligent diagnosis method based on cold end offline data according to claim 2 is characterized in that: The server verifies the prediction result, and the verification is to compare the prediction result with the corresponding offline data set.
4. The intelligent diagnosis method based on cold end offline data according to claim 3 is characterized in that: Through verification on the server side, the prediction results are mapped to the offline data set on the cold side, and the offline data set on the cold side is used for machine learning of the intelligent diagnosis model.
5. The intelligent diagnosis method based on cold end offline data according to claim 1 is characterized in that: The cold end offline diagnosis result output by the server includes a verification prediction result, which is a prediction result obtained by the server through a historical offline data set.
6. The intelligent diagnosis method based on cold end offline data according to claim 2 is characterized in that: The offline data set includes Y1 at time point t: offline flow of circulating water; Y2: heat load of condenser; and Y3: efficiency of cooling tower.
7. The intelligent diagnosis method based on cold end offline data according to claim 6 is characterized in that: The intelligent diagnosis model is Among them, c1, c2, c3 are constant terms, ij,k is the lag term coefficient, it Error term.
8. The intelligent diagnosis method based on cold end offline data according to claim 7 is characterized in that: The prediction result at time t+1 is the prediction value calculated using the offline data sets at time t and time t-1.
9. The intelligent diagnosis method based on cold end offline data according to claim 8, characterized in that: The offline data set at time t+1 is used to remove abnormal points in the offline data sets at time t and time t-1.
10. The intelligent diagnosis method based on cold end offline data according to claim 8, characterized in that: The offline data set at time t+1 is used to correct the abnormal points in the offline data sets at time t and time t-1.