Backwater optimization system of TCA cooler

Abstract

The invention provides a TCA cooler backwater optimization system, which relates to the technical field of cooler backwater optimization and comprises a data acquisition module, a data preprocessing module, a data storage module, a data analysis module, an intelligent optimization module and a prediction maintenance module. The method comprises the following steps: acquiring temperature data, water flow data, pipeline pressure data, regulating valve opening data and equipment vibration data of a main water return pipeline and a parallel branch, and performing optimization deviation processing and feature extraction processing on the data in a historical acquisition period to obtain an optimization deviation data set and a vibration feature value; further obtaining a deep analysis data set; a correlation coefficient matrix is constructed, a dynamic adjustment index is obtained, and dynamic switching of water return paths between the main water return pipeline and the parallel branches is achieved; constructing a multi-dimensional data prediction maintenance model, generating a prediction equipment fault risk index, obtaining an equipment risk assessment level, and generating equipment risk early warning information according to the equipment risk assessment level; and the operation economy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of cooler return water optimization, and specifically to a TCA cooler return water optimization system. Background Art

[0002] For a gas turbine combined cycle with a turbine cooling air (TCA) cooler system, the return water position after pumping water from the waste heat boiler for heat exchange is usually set at the outlet of the high-pressure economizer; The main defects are as follows: temperature imbalance: when the pumping water flow rate to the TCA cooler increases and the return water temperature is significantly lower than the working medium temperature at the outlet of the high-pressure economizer, the direct entry of the return water into the steam drum will cause insufficient heat absorption in the high-pressure part, resulting in a decrease in system output; redundant equipment selection: the heat not absorbed by the high-pressure part is transferred to the medium and low-pressure equipment, leading to over-sizing of the medium and low-pressure equipment and increasing construction and operation costs; poor economy: the system cannot adapt to the fluctuations in the pumping water flow rate to the TCA cooler and requires frequent manual adjustment, affecting the operation efficiency of the power plant.

[0003] In order to solve the above defects, therefore, a TCA cooler return water optimization system is provided now. Summary of the Invention

[0004] In order to solve the above technical problems, the purpose of the present invention is to provide a TCA cooler return water optimization system.

[0005] In order to achieve the above purpose, the present invention provides the following technical solution: a TCA cooler return water optimization system, including: a data acquisition module, a data preprocessing module, a data storage module, a data analysis module, an intelligent optimization module, and a predictive maintenance module; The data acquisition module acquires temperature data, water flow rate data, pipeline pressure data, regulating valve opening data, and equipment vibration data of the main return water pipeline and the parallel branches; The data preprocessing module preprocesses the acquired data to obtain preprocessed temperature data, preprocessed water flow rate data, preprocessed pipeline pressure data, preprocessed regulating valve opening data, and preprocessed equipment vibration data; The data storage module stores the preprocessed data to obtain historical temperature data, historical water flow rate data, historical pipeline pressure data, historical regulating valve opening data, and historical equipment vibration data; The data analysis module performs optimization deviation processing on the historical temperature data, historical water flow rate data, historical pipeline pressure data, and historical regulating valve opening data respectively to obtain an optimization deviation data set; performs feature extraction processing on the historical equipment vibration data to obtain vibration characteristic values, and further obtains a depth analysis data set; The intelligent optimization module constructs a correlation coefficient matrix, obtains a dynamic adjustment index, and realizes the dynamic switching of the return water path between the main return water pipeline and the parallel branches; The predictive maintenance module constructs a multi-dimensional data predictive maintenance model based on the in-depth analysis dataset, generates a predictive device failure risk index, obtains the device risk assessment level based on the predictive device failure risk index, and further generates device risk warning information.

[0006] According to one preferred embodiment of the present invention, the process of collecting temperature data, water flow data, pipeline pressure data, regulating valve opening data, and device vibration data of the main return water pipeline and the parallel branch includes: Set up a data acquisition device, which is composed of several temperature sensing units, water flow sensing units, pressure sensing units, regulating valve opening sensing units, and vibration sensing units, and set an acquisition period, which is composed of several acquisition moments; collect the temperature data, water flow data, pipeline pressure data, regulating valve opening data, and device vibration data of the main return water pipeline and the parallel branch according to the acquisition period.

[0007] According to one preferred embodiment of the present invention, the process of preprocessing the collected data includes: Based on data cleaning technology, perform data cleaning on the temperature data, water flow data, pipeline pressure data, regulating valve opening data, and device vibration data; Perform filtering processing on the temperature data, water flow data, pipeline pressure data, regulating valve opening data, and device vibration data after data cleaning; Convert the temperature data, water flow data, pipeline pressure data, regulating valve opening data, and device vibration data after filtering processing into digital quantity data through an analog-to-digital converter; Perform normalization processing on the temperature data, water flow data, pipeline pressure data, regulating valve opening data, and device vibration data after the data format conversion processing; record the data after normalization processing as preprocessed temperature data, preprocessed water flow data, preprocessed pipeline pressure data, preprocessed regulating valve opening data, and preprocessed device vibration data.

[0008] According to one preferred embodiment of the present invention, the process of storing the preprocessed data includes: Set up a real-time database and a historical database, and send the preprocessed temperature data, preprocessed water flow data, preprocessed pipeline pressure data, preprocessed regulating valve opening data, and preprocessed device vibration data in the current acquisition period to the real-time database for storage; Obtain the pre - treatment temperature data, pre - treatment water flow rate data, pre - treatment pipeline pressure data, pre - treatment regulating valve opening data, and pre - treatment equipment vibration data within the next acquisition cycle, and send them to the real - time database; send the pre - treatment temperature data, pre - treatment water flow rate data, pre - treatment pipeline pressure data, pre - treatment regulating valve opening data, and pre - treatment equipment vibration data collected in the previous acquisition cycle to the historical database for storage, and mark them as historical temperature data, historical water flow rate data, historical pipeline pressure data, historical regulating valve opening data, and historical equipment vibration data.

[0009] According to one preferred embodiment of the present invention, the process of performing optimization deviation processing includes: Obtain several groups of historical temperature data, historical water flow rate data, historical pipeline pressure data, and historical regulating valve opening data within the acquisition cycle; For several groups of historical temperature data within the acquisition cycle in the historical database , perform marking where \(i\) is a natural number; According to the marked several groups of historical temperature data within the acquisition cycle , obtain the optimized temperature data The calculation formula is: ; where \(\alpha\) is the deviation coefficient of the temperature data; For several groups of historical water flow rate data within the acquisition cycle in the historical database , perform marking where \(i\) is a natural number; According to the marked several groups of historical water flow rate data within the acquisition cycle , obtain the optimized water flow rate data The calculation formula is: ; where \(\beta\) is the deviation coefficient of the water flow rate data; For several groups of historical pipeline pressure data within the acquisition cycle in the historical database , perform marking where \(i\) is a natural number; According to the marked several groups of historical pipeline pressure data within the acquisition cycle , obtain the optimized pipeline pressure data The calculation formula is ; where \(\gamma\) is the deviation coefficient of the pipeline pressure data; For several groups of historical regulating valve opening data within the acquisition cycle in the historical database , perform marking where \(i\) is a natural number; According to the historical pipeline pressure data within several marked acquisition cycles , obtain the optimized meteorological data The calculation formula is: ; where is the deviation coefficient of the pipeline pressure data; The optimized temperature data within the same acquisition cycle , the optimized water flow rate data , the optimized pipeline pressure data , and the optimized meteorological data are denoted as the optimized deviation data set.

[0010] According to one preferred embodiment of the present invention, the process of feature extraction processing includes: Obtain the historical equipment vibration data; For the historical equipment vibration data within several acquisition cycles in the historical database , perform marking as a natural number; According to the historical equipment vibration data within several marked acquisition cycles , obtain the vibration characteristic value The calculation formula is: ; where is the deviation coefficient of the equipment vibration data; According to the optimized deviation data set and the vibration characteristic value within the same acquisition cycle , obtain the in-depth analysis optimization data set; obtain the in-depth analysis optimization data set for each acquisition cycle within the historical acquisition cycle.

[0011] According to one preferred embodiment of the present invention, the process of constructing a correlation coefficient matrix and then obtaining a dynamic adjustment index to realize the dynamic switching of the return water path between the main return water pipeline and the parallel branch includes: Obtain the optimized temperature data , the optimized water flow rate data , the optimized pipeline pressure data , the optimized meteorological data and the vibration characteristic value in the in-depth analysis optimization data set; The process of constructing a correlation coefficient matrix includes: According to the optimized temperature data , the optimized water flow rate data , the optimized pipeline pressure data , the optimized meteorological data and the vibration characteristic value , calculate the Pearson correlation coefficient and Spearman correlation coefficient between every two columns of data in sequence, and obtain a Spearman correlation coefficient matrix with a dimension of , denoted as and a Pearson correlation coefficient matrix, denoted as ; According to the Pearson correlation coefficient matrix and the Spearman correlation coefficient matrix , construct a correlation coefficient matrix, denoted as , and the correlation coefficient matrix is: ; where , are weight parameters; Based on matrix numerical calculation, perform eigenvalue decomposition on the correlation coefficient matrix , and obtain eigenvalues, denoted as ; According to the eigenvalues , obtain a dynamic adjustment index , and the dynamic adjustment index is: ; where is a matrix characteristic parameter; Preset the dynamic adjustment index range interval ; When the dynamic adjustment index is greater than , the device automatically triggers a return water path switching instruction; When the dynamic adjustment index is within , the device issues a warning prompt and sets a start timer; When the dynamic adjustment index is less than and remains stable within the end of the start timer, the device gradually returns to the default return water path.

[0012] According to one preferred embodiment of the present invention, the process of constructing a multi-dimensional data prediction and maintenance model includes: Obtain several groups of in-depth analysis and optimization data sets; preset a standard in-depth analysis and optimization data set; Group and label several groups of in-depth analysis and optimization data sets, denoted as being a natural number; Take groups of several groups of in-depth analysis and optimization data sets and the standard in-depth analysis and optimization data set as sample data, and is less than natural numbers, and using the sample data, obtain the mean of the sample data, denoted as the sample set; use the remaining several groups of in-depth analysis and optimization data sets and the standard in-depth analysis and optimization data set as the test set; According to the sample set and the test set, form a training sample set; based on the convolutional neural network, construct a standard prediction model; And input the training sample set into the standard prediction model to train the standard prediction model, and obtain the trained standard prediction model, and denote the trained standard prediction model as the multi-dimensional data prediction and maintenance model; According to the multi-dimensional data prediction and maintenance model, generate the predicted equipment failure risk index under the current environmental factor conditions , the predicted equipment failure risk index is: ; wherein, is the weight coefficient of the optimized temperature data and ; is the weight coefficient of the optimized water flow data and ; is the weight coefficient of the optimized pipeline pressure data and ; is the weight coefficient of the optimized meteorological data and ; is the weight coefficient of the vibration eigenvalue and ; is the standard deviation value of the multi-dimensional data prediction and maintenance model.

[0013] According to one preferred embodiment of the present invention, the process of obtaining the equipment risk assessment level according to the predicted equipment failure risk index and generating the equipment risk warning information according to the equipment risk assessment level includes: Preset the standard failure risk index ; If the predicted equipment failure risk index , then there is no failure risk at the corresponding measurement points of the main return water pipeline, the parallel branch, the first regulating valve on the main return water pipeline, and the second regulating valve on the parallel branch under the current environmental factor conditions; If the predicted equipment failure risk index , there is a risk of failure for the corresponding measuring points of the main return water pipeline, parallel branches, the first regulating valve on the main return water pipeline, and the second regulating valve on the parallel branches under the current environmental factor conditions, and an equipment risk assessment level is generated; the equipment risk assessment level includes a first-level equipment risk assessment, a second-level equipment risk assessment, a third-level equipment risk assessment, and a fourth-level equipment risk assessment; When there is a first-level equipment risk assessment, a main return water pipeline equipment risk warning message is generated; when there is a second-level equipment risk assessment, a parallel branch equipment risk warning message is generated; when there is a third-level equipment risk assessment, a first regulating valve equipment risk warning message is generated; when there is a fourth-level equipment risk assessment, a second regulating valve equipment risk warning message is generated.

[0014] The present invention further provides a computer-readable storage medium storing a computer program that can be executed by a processor to implement the above-mentioned TCA cooler return water optimization system.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: collecting the process parameter data and equipment vibration data of the main return water pipeline and parallel branches, and performing preprocessing to obtain preprocessed process parameter data and preprocessed equipment vibration data, and then storing them to obtain historical process parameter data and historical equipment vibration data, and performing in-depth analysis to obtain an in-depth analysis data set; constructing a dynamic optimization model to obtain a dynamic adjustment index to realize the dynamic switching of the return water path between the main return water pipeline and parallel branches; constructing a multi-dimensional data prediction and maintenance model to generate a predicted equipment failure risk index, obtaining an equipment risk assessment level, and generating an equipment risk warning message according to the equipment risk assessment level; improving the economic efficiency of operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings.

[0017] Figure 1 It is a step schematic diagram of a TCA cooler return water optimization system.

[0018] Figure 2 It is a flow schematic diagram of a TCA cooler return water optimization system.

[0019] Figure 3 It is a module schematic diagram of a TCA cooler return water optimization system.

[0020] Figure 4It is a schematic diagram of the equipment for an optimized system of the return water of a TCA cooler. Detailed implementation mode

[0021] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be described in detail below. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other implementation manners obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope protected by the present invention.

[0022] As Figure 1 、 Figure 3 shown, an optimized system of the return water of a TCA cooler includes: a data acquisition module, a data preprocessing module, a data storage module, a data analysis module, an intelligent optimization module and a predictive maintenance module; The data acquisition module acquires the temperature data, water flow data, pipeline pressure data, regulating valve opening data and equipment vibration data of the main return water pipeline and the parallel branches; The data preprocessing module preprocesses the acquired data to obtain preprocessed temperature data, preprocessed water flow data, preprocessed pipeline pressure data, preprocessed regulating valve opening data and preprocessed equipment vibration data; The data storage module stores the preprocessed data to obtain historical temperature data, historical water flow data, historical pipeline pressure data, historical regulating valve opening data and historical equipment vibration data; The data analysis module respectively performs optimization deviation processing on the historical temperature data, historical water flow data, historical pipeline pressure data and historical regulating valve opening data to obtain an optimized deviation data set; performs feature extraction processing on the historical equipment vibration data to obtain vibration characteristic values, and further obtains a deep analysis data set; The intelligent optimization module constructs a correlation coefficient matrix, obtains a dynamic adjustment index, and realizes the dynamic switching of the return water path between the main return water pipeline and the parallel branches; The predictive maintenance module constructs a multi-dimensional data predictive maintenance model according to the deep analysis data set, generates a predictive equipment failure risk index, obtains an equipment risk assessment level according to the predictive equipment failure risk index, and further generates equipment risk warning information.

[0023] It should be further noted that in the specific implementation process, the specific process of acquiring the temperature data, water flow data, pipeline pressure data, regulating valve opening data and equipment vibration data of the main return water pipeline and the parallel branches includes: A data acquisition device is set up. The data acquisition device consists of several temperature sensing units, water flow sensing units, pressure sensing units, regulating valve opening sensing units, and vibration sensing units, and an acquisition period is set. The acquisition period consists of several acquisition times. The temperature sensing unit, water flow sensing unit, pressure sensing unit, regulating valve opening sensing unit, and vibration sensing unit are respectively set at the corresponding measuring points of the main return water pipeline, parallel branch, first regulating valve on the main return water pipeline, and second regulating valve on the parallel branch; and the temperature data, water flow data, pipeline pressure data, regulating valve opening data, and equipment vibration data of the corresponding measuring points are acquired according to the acquisition period. It should be further noted that in the specific implementation process, the specific process of preprocessing the temperature data, water flow data, pipeline pressure data, regulating valve opening data, and equipment vibration data includes: Obtain the temperature data, water flow data, pipeline pressure data, regulating valve opening data, and equipment vibration data. Based on the data cleaning technology, perform data cleaning on the temperature data, water flow data, pipeline pressure data, regulating valve opening data, and equipment vibration data. For the temperature data, water flow data, pipeline pressure data, and regulating valve opening data after data cleaning, use a first-order low-pass filter and perform filtering processing through the discretized difference equation. Adjust the time constant to control the cut-off frequency and remove the high-frequency noise in the corresponding data; it should be further noted that when adjusting the time constant, avoid data distortion caused by adjusting the time constant. For the equipment vibration data after data cleaning, perform median filtering for filtering processing. Select an odd-length window and replace the original data point with the median value of the data within the window to effectively remove impulse noise; it should be further noted that the equipment vibration data is easily interfered by impulse noise. Convert the temperature data, water flow data, pipeline pressure data, regulating valve opening data, and equipment vibration data after filtering processing into digital quantity data through an analog-to-digital converter, and perform data format conversion according to the range of the corresponding sensing unit and the input range of the analog-to-digital converter for subsequent data analysis. And perform normalization processing on the temperature data, water flow data, pipeline pressure data, regulating valve opening data, and equipment vibration data after the data format conversion processing, and normalize them to the range, which is convenient for subsequent unified analysis and improving the algorithm performance. Record the data after normalization processing as preprocessed temperature data, preprocessed water flow data, preprocessed pipeline pressure data, preprocessed regulating valve opening data, and preprocessed equipment vibration data.

[0024] It should be further noted that in the specific implementation process, the specific process of storing the preprocessed data includes: Set up a real-time database and a historical database, obtain the preprocessed temperature data, preprocessed water flow data, preprocessed pipeline pressure data, preprocessed regulating valve opening data, and preprocessed equipment vibration data within the current collection cycle, and send the preprocessed temperature data, preprocessed water flow data, preprocessed pipeline pressure data, preprocessed regulating valve opening data, and preprocessed equipment vibration data within the current collection cycle to the real-time database for storage; Obtain the preprocessed temperature data, preprocessed water flow data, preprocessed pipeline pressure data, preprocessed regulating valve opening data, and preprocessed equipment vibration data within the next collection cycle, and send them to the real-time database; when the real-time database receives the preprocessed temperature data, preprocessed water flow data, preprocessed pipeline pressure data, preprocessed regulating valve opening data, and preprocessed equipment vibration data within the next collection cycle, send the preprocessed temperature data, preprocessed water flow data, preprocessed pipeline pressure data, preprocessed regulating valve opening data, and preprocessed equipment vibration data collected in the previous collection cycle to the historical database for storage, and mark them as historical temperature data, historical water flow data, historical pipeline pressure data, historical regulating valve opening data, and historical equipment vibration data.

[0025] It should be further noted that in the specific implementation process, the specific process of optimizing the deviation of historical temperature data, historical water flow data, historical pipeline pressure data, and historical regulating valve opening data includes: Obtain several groups of historical temperature data, historical water flow data, historical pipeline pressure data, historical regulating valve opening data, and historical equipment vibration data within the collection cycle; For several groups of historical temperature data within the collection cycle in the historical database , make a mark which is a natural number; According to the marked several groups of historical temperature data within the collection cycle , obtain the optimized temperature data The calculation formula is: ; where is the deviation coefficient of the temperature data; For several groups of historical water flow data within the collection cycle in the historical database , make a mark which is a natural number; According to the marked several groups of historical water flow data within the collection cycle , obtain the optimized water flow data The calculation formula is: ; among which, is the deviation coefficient of water flow data; For several groups of historical pipeline pressure data within the acquisition period in the historical database , conduct marking where \(n\) is a natural number; According to the marked several groups of historical pipeline pressure data within the acquisition period , obtain the optimized pipeline pressure data The calculation formula is ; among which, is the deviation coefficient of pipeline pressure data; For several groups of historical regulating valve opening data within the acquisition period in the historical database , conduct marking where \(n\) is a natural number; According to the marked several groups of historical pipeline pressure data within the acquisition period , obtain the optimized meteorological data The calculation formula is: ; among which, is the deviation coefficient of pipeline pressure data; According to the optimized temperature data, optimized water flow data, optimized pipeline pressure data, and optimized meteorological data within the same acquisition period , , , denote as the optimized deviation data set; It should be further noted that in the specific implementation process, the specific process of feature processing for historical equipment vibration data includes: For several groups of historical equipment vibration data within the acquisition period in the historical database , conduct marking where \(n\) is a natural number; According to the marked several groups of historical equipment vibration data within the acquisition period , obtain the characteristic equipment vibration data The calculation formula is: ; among which, is the deviation coefficient of equipment vibration data; According to the optimized deviation data set and the characteristic equipment vibration data within the same acquisition period , obtain the in-depth analysis optimized data set; Similarly, obtain the in-depth analysis optimized data set for each acquisition period within the historical acquisition period.

[0026] It should be further noted that in the specific implementation process, the specific process of constructing the correlation coefficient matrix, obtaining the dynamic adjustment index, and realizing the dynamic switching of the return water path between the main return water pipeline and the parallel branch includes: Obtain the optimized temperature data in the in-depth analysis and optimization dataset , the optimized water flow data , the optimized pipeline pressure data , the optimized meteorological data and the vibration eigenvalue ; Based on the calculation of the Pearson correlation coefficient, obtain the Pearson correlation coefficient between the corresponding data, and based on the calculation of the Spearman correlation coefficient, obtain the Spearman correlation coefficient between the corresponding data; The specific process of constructing the correlation coefficient matrix according to the Pearson correlation coefficient and the Spearman correlation coefficient includes: According to the optimized temperature data , the optimized water flow data , the optimized pipeline pressure data , the optimized meteorological data and the vibration eigenvalue , calculate the Pearson correlation coefficient between every two columns of data in turn to obtain the Pearson correlation coefficient matrix with the dimension of , denoted as ; According to the optimized temperature data , the optimized water flow data , the optimized pipeline pressure data , the optimized meteorological data and the vibration eigenvalue , calculate the Spearman correlation coefficient between every two columns of data in turn to obtain the Spearman correlation coefficient matrix with the dimension of , denoted as ; According to the Pearson correlation coefficient matrix and the Spearman correlation coefficient matrix , construct the correlation coefficient matrix, denoted as , and the correlation coefficient matrix is: ; where , are weight parameters; Based on matrix numerical calculation, perform eigenvalue decomposition on the correlation coefficient matrix to obtain the eigenvalues, denoted as ; According to the eigenvalues , obtain the dynamic adjustment index , the dynamic adjustment index is: ; where is the matrix characteristic parameter, which can be adjusted according to the actual situation to adapt to the characteristics of the data and the requirements of the system; preset dynamic adjustment index range interval ; When the dynamic adjustment index is greater than , the device automatically triggers a return water path switching instruction and prepares to switch the return water path from the current state to a more appropriate path; the current return water passes through the main return water pipeline, and at this time, it may be necessary to open the second regulating valve of the parallel branch to change the flow direction of the return water; When the dynamic adjustment index is within , the device issues a warning prompt, sets a start timer, and continuously monitors the change trend of the dynamic adjustment index during the operation of the start timer. If, at the end of the start timer, the dynamic adjustment index rises and reaches or exceeds , the return water path switching operation is executed; if the dynamic adjustment index drops or remains stable, the system state is continuously observed and no path switching is performed; When the dynamic adjustment index is less than and remains stable within the end of the start timer, the device gradually returns to the default return water path, that is, closes the second regulating valve of the parallel branch to make the return water mainly flow through the main return water pipeline again; For example, as Figure 4 shows, the device includes a main return water pipeline, a parallel branch, a first regulating valve arranged on the main return water pipeline, and a second regulating valve arranged on the parallel branch; the parallel branch is connected between the high-pressure secondary economizer and the high-pressure last-stage economizer; The preset initial state of the device operation is: the first regulating valve is open, the second regulating valve is closed, and the return water returns to the outlet of the high-pressure economizer through the main return water pipeline; Flow increase scenario: When the pumping flow of the device increases and causes the return water temperature to drop, the temperature sensing unit triggers a signal to close the first regulating valve and open the second regulating valve. The low-temperature return water enters the high-pressure economizer stage through the parallel branch and is further heated by the high-pressure last-stage economizer to increase the output of the high-pressure part; Flow decrease scenario: When the pumping flow of the device decreases and the return water temperature rises, the temperature sensing unit triggers a signal to close the second regulating valve and open the first regulating valve, and the system returns to the default path.

[0027] It should be further noted that, in the specific implementation process, the specific process of constructing the multi-dimensional data prediction and maintenance model includes: Obtain several groups of in-depth analysis and optimization data sets; Preset a standard in-depth analysis and optimization data set; Group and label several groups of in-depth analysis and optimization data sets, denoted as where is a natural number; Take groups of several groups of in-depth analysis and optimization data sets and the standard in-depth analysis and optimization data set as sample data, and is a natural number less than and use the said sample data to obtain the mean value of the sample data, denoted as the sample set; Take the remaining several groups of in-depth analysis and optimization data sets and the standard in-depth analysis and optimization data set as the test set; According to the said sample set and the test set, form a training sample set; Based on the convolutional neural network, construct a standard prediction model; And input the training sample set into the standard prediction model to train the standard prediction model, obtain the trained standard prediction model, and denote the trained standard prediction model as the multi-dimensional data prediction and maintenance model; According to the multi-dimensional data prediction and maintenance model, generate the predicted equipment failure risk index under the current environmental factor conditions , the predicted equipment failure risk index is: ; where, is the weight coefficient of the optimized temperature data and ; is the weight coefficient of the optimized water flow data and ; is the weight coefficient of the optimized pipeline pressure data and ; is the weight coefficient of the optimized meteorological data and ; is the weight coefficient of the vibration eigenvalue and ; is the standard deviation value of the multi-dimensional data prediction and maintenance model.

[0028] It should be further noted that in the specific implementation process, the specific process of obtaining the equipment risk assessment level according to the predicted equipment failure risk index and generating the equipment risk warning information according to the equipment risk assessment level includes: Preset a standard failure risk index ; If the predicted equipment failure risk index , there is no risk of failure in the corresponding measuring points of the main return water pipeline, parallel branch, first regulating valve on the main return water pipeline and second regulating valve on the parallel branch under the current environmental factor conditions; If the predicted equipment failure risk index , there is a risk of failure in the corresponding measuring points of the main return water pipeline, parallel branch, first regulating valve on the main return water pipeline and second regulating valve on the parallel branch under the current environmental factor conditions, and an equipment risk assessment level is generated; It should be further explained that the equipment risk assessment level includes primary equipment risk assessment, secondary equipment risk assessment, tertiary equipment risk assessment and quaternary equipment risk assessment; If there is a primary equipment risk assessment, a main return water pipeline equipment risk warning message is generated; If there is a secondary equipment risk assessment, a parallel branch equipment risk warning message is generated; If there is a tertiary equipment risk assessment, a first regulating valve equipment risk warning message is generated; If there is a quaternary equipment risk assessment, a second regulating valve equipment risk warning message is generated; So as to facilitate the staff to maintain the operating equipment in a timely manner.

[0029] The above embodiments are only used to illustrate the technical method of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A TCA cooler return water optimization system, characterized in that, Including: A data acquisition module that acquires temperature data, water flow data, pipeline pressure data, regulating valve opening data, and equipment vibration data of the main return water pipeline and parallel branches; A data preprocessing module that preprocesses the acquired data to obtain preprocessed temperature data, preprocessed water flow data, preprocessed pipeline pressure data, preprocessed regulating valve opening data, and preprocessed equipment vibration data; A data storage module that stores the preprocessed data to obtain historical temperature data, historical water flow data, historical pipeline pressure data, historical regulating valve opening data, and historical equipment vibration data; A data analysis module that performs optimization deviation processing on historical temperature data, historical water flow data, historical pipeline pressure data, and historical regulating valve opening data respectively to obtain an optimized deviation data set; performs feature extraction processing on historical equipment vibration data to obtain vibration eigenvalues, and further obtains a deep analysis data set; An intelligent optimization module that constructs a correlation coefficient matrix, obtains a dynamic adjustment index, and realizes the dynamic switching of the return water path between the main return water pipeline and parallel branches; A predictive maintenance module that constructs a multi-dimensional data predictive maintenance model based on the deep analysis data set, generates a predicted equipment failure risk index, obtains an equipment risk assessment level based on the predicted equipment failure risk index, and further generates an equipment risk warning message.

2. The optimized system for the return water of a TCA cooler according to claim 1, wherein The process of acquiring temperature data, water flow data, pipeline pressure data, regulating valve opening data, and equipment vibration data of the main return water pipeline and parallel branches includes: Setting a data acquisition device, which is composed of several temperature sensing units, water flow sensing units, pressure sensing units, regulating valve opening sensing units, and vibration sensing units, and setting an acquisition period, which is composed of several acquisition moments; acquiring temperature data, water flow data, pipeline pressure data, regulating valve opening data, and equipment vibration data of the main return water pipeline and parallel branches according to the acquisition period.

3. The optimized system for the return water of the TCA cooler according to claim 2, characterized in that, The process of preprocessing the acquired data includes: Based on data cleaning technology, cleaning the temperature data, water flow data, pipeline pressure data, regulating valve opening data, and equipment vibration data; Performing filtering processing on the temperature data, water flow data, pipeline pressure data, regulating valve opening data, and equipment vibration data after data cleaning; Converting the temperature data, water flow data, pipeline pressure data, regulating valve opening data, and equipment vibration data after filtering processing into digital quantity data through an analog-to-digital converter; Performing normalization processing on the temperature data, water flow data, pipeline pressure data, regulating valve opening data, and equipment vibration data after the data format conversion processing; and recording the normalized data as preprocessed temperature data, preprocessed water flow data, preprocessed pipeline pressure data, preprocessed regulating valve opening data, and preprocessed equipment vibration data.

4. The optimized system for the return water of a TCA cooler according to claim 3, characterized in that, The process of storing the preprocessed data includes: Set up a real-time database and a historical database, and send the preprocessed temperature data, preprocessed water flow data, preprocessed pipeline pressure data, preprocessed regulating valve opening data, and preprocessed equipment vibration data within the current collection period to the real-time database for storage; Obtain the preprocessed temperature data, preprocessed water flow data, preprocessed pipeline pressure data, preprocessed regulating valve opening data, and preprocessed equipment vibration data within the next collection period, and send them to the real-time database; send the preprocessed temperature data, preprocessed water flow data, preprocessed pipeline pressure data, preprocessed regulating valve opening data, and preprocessed equipment vibration data collected within the previous collection period to the historical database for storage, and mark them as historical temperature data, historical water flow data, historical pipeline pressure data, historical regulating valve opening data, and historical equipment vibration data.

5. The optimized return water system of a TCA cooler according to claim 4, characterized in that, The process of performing optimization deviation processing includes: Obtain several groups of historical temperature data, historical water flow data, historical pipeline pressure data, and historical regulating valve opening data within the collection period; For several groups of historical temperature data within the acquisition cycles in the historical database , perform marking as natural numbers; According to the historical temperature data within several marked acquisition cycles , the optimized temperature data is calculated by the formula: ; wherein, is the deviation coefficient of the temperature data; For several sets of historical water flow data within the collection periods in the historical database , perform marking as natural numbers; According to the historical water flow data within several marked acquisition cycles , the optimized water flow data is calculated by the formula: ; wherein, is the deviation coefficient of the water flow rate data; For several groups of historical pipeline pressure data within the acquisition period in the historical database , perform marking as natural numbers; According to the historical pipeline pressure data within several marked acquisition cycles , the optimized pipeline pressure data is calculated by the formula ; wherein, is the deviation coefficient of the pipeline pressure data; For several groups of historical regulating valve opening data within the collection periods in the historical database , perform marking as natural numbers; According to the historical pipeline pressure data within several marked acquisition periods , the optimized meteorological data is calculated by the formula: ; wherein, is the deviation coefficient of the pipeline pressure data; Optimized temperature data within the same acquisition period , optimized water flow rate data , optimized pipeline pressure data , optimized meteorological data are recorded as the optimized deviation data set.

6. The optimized system for the return water of the TCA cooler according to claim 5, characterized in that, The process of performing feature extraction processing includes: Obtain historical equipment vibration data; For several groups of historical equipment vibration data within the acquisition periods in the historical database , perform marking as natural numbers; According to the historical device vibration data within several marked acquisition periods , the vibration characteristic value is calculated by the formula: ; wherein, is the deviation coefficient of the device vibration data; According to the optimized deviation data set and vibration characteristic values within the same acquisition period , an in-depth analysis optimized data set is obtained; an in-depth analysis optimized data set for each acquisition period within the historical acquisition period is obtained.

7. The optimized system for the return water of the TCA cooler according to claim 6, characterized in that, The process of constructing a correlation coefficient matrix and then obtaining a dynamic adjustment index to achieve dynamic switching of the return water path between the main return water pipeline and the parallel branch includes: Obtain the optimized temperature data in the in-depth analysis optimization dataset , the optimized water flow rate data , the optimized pipeline pressure data , the optimized meteorological data and the vibration eigenvalue ; The process of constructing a correlation coefficient matrix includes: According to the optimized temperature data , the optimized water flow rate data , the optimized pipeline pressure data , the optimized meteorological data and the vibration eigenvalue , calculate the Pearson correlation coefficient and the Spearman correlation coefficient between every two columns of data in sequence, and obtain the Spearman correlation coefficient matrix with the dimension of , denoted as and the Pearson correlation coefficient matrix, denoted as ; According to the Pearson correlation coefficient matrix and the Spearman correlation coefficient matrix , a correlation coefficient matrix is constructed and denoted as . The correlation coefficient matrix is as follows: ; wherein, and are weight parameters; Based on matrix numerical calculation, for the said correlation coefficient matrix perform eigenvalue decomposition to obtain eigenvalues, denoted as ; According to the eigenvalue , a dynamic adjustment index is obtained , and the dynamic adjustment index is as follows: ; wherein, is the matrix characteristic parameter; Preset dynamic adjustment index range interval ; When dynamically adjusting the index is greater than the device automatically triggers a return water path switching instruction; When the dynamic adjustment index is at the device issues a warning prompt and sets a start timer; When the dynamic adjustment index is less than and remains stable within the end of the start timer, the device gradually returns to the default return water path.

8. The optimized system for the return water of a TCA cooler according to claim 7, wherein, The process of constructing a multi-dimensional data prediction and maintenance model includes: Obtain several groups of in-depth analysis and optimization data sets; preset a standard in-depth analysis and optimization data set; Group and label several groups of deep analysis optimization data sets, denoted as is a natural number; Put Several groups of in-depth analysis optimization data sets and standard in-depth analysis optimization data sets are used as sample data, and is a natural number less than Using the sample data, the mean value of the sample data is obtained, denoted as the sample set; the remaining several groups of in-depth analysis optimization data sets and standard in-depth analysis optimization data sets are used as the test set; According to the sample set and the test set, form a training sample set; based on the convolutional neural network, construct a standard prediction model; And input the training sample set into the standard prediction model, train the standard prediction model, obtain the trained standard prediction model, and record the trained standard prediction model as a multi-dimensional data prediction and maintenance model; Generate a predicted equipment failure risk index under the current environmental factor conditions according to the multi-dimensional data prediction and maintenance model , the predicted equipment failure risk index is as follows: ； Among them, is the weight coefficient for optimizing temperature data and ; is the weight coefficient for optimizing water flow data and ; is the weight coefficient for optimizing pipeline pressure data and ; is the weight coefficient for optimizing meteorological data and ; is the weight coefficient for the vibration eigenvalue and ; is the standard deviation value of the multi-dimensional data prediction and maintenance model.

9. The optimized system for the return water of the TCA cooler according to claim 8, wherein, The process of obtaining the equipment risk assessment level according to the predicted equipment failure risk index and generating equipment risk warning information according to the equipment risk assessment level includes: Preset standard fault risk index ; If the predicted equipment failure risk index , then there is no failure risk at the corresponding measuring points of the main return water pipeline, parallel branch, first regulating valve on the main return water pipeline, and second regulating valve on the parallel branch under the current environmental factor conditions; If the predicted equipment failure risk index is reached, there are failure risks at the corresponding measuring points of the main return water pipeline, the parallel branch, the first regulating valve on the main return water pipeline, and the second regulating valve on the parallel branch under the current environmental factor conditions, and an equipment risk assessment level is generated; the equipment risk assessment level includes a first-level equipment risk assessment, a second-level equipment risk assessment, a third-level equipment risk assessment, and a fourth-level equipment risk assessment; If there is a first-level equipment risk assessment, generate a main return water pipeline equipment risk warning information; if there is a second-level equipment risk assessment, generate a parallel branch equipment risk warning information; if there is a third-level equipment risk assessment, generate a first regulating valve equipment risk warning information; if there is a fourth-level equipment risk assessment, generate a second regulating valve equipment risk warning information.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and the computer program can be executed by a processor to implement a TCA cooler return water optimization system according to any one of claims 1-9 above.

Images