Install an external detachable one-way valve system on the oil pump outlet of the small engine oil tank

Abstract

and a tube connecting the discharging opening of the oil drain plug, wherein the tube has a check valve in it and is closed by the pump , and the tube which connects the pump to the oil drain plug, and the pump is closed by the pump .

Description

Technical Field

[0001] The invention relates to the technical field of steam turbines, in particular to a system for adding an externally mounted detachable one-way valve to an outlet of an oil pump in an oil tank of a small engine. Background Art

[0002] The small unit main oil tank is mainly used to store the working fluid required by the hydraulic system. It also has the functions of heat dissipation, impurity filtering and air bubble removal in the liquid. The main function of the small unit main oil pump is to supply oil to the various bearings of the unit, lubricate and cool the bearings, and supply stable and sufficient pressure oil to the speed control system and protective devices. It is one of the important auxiliary components of the steam turbine and is installed inside the small unit main oil tank. When the steam turbine is at full speed, the main oil pump provides all the lubricating oil and safety oil for the unit. During the start and stop of the steam turbine, the unit lubricating oil and safety oil are provided by the auxiliary AC lubricating oil pump and high-pressure starting oil pump. Using the main oil pump can not only save plant electricity, but also prevent the risk of oil shortage and tile burning due to power failure of the AC lubricating pump.

[0003] The main oil pump of a small engine is usually connected with a one-way valve. When the oil pump is started, the air in the pump is quickly discharged through the one-way exhaust valve to quickly build up oil pressure. The one-way valve is inside the main oil tank of the small engine. Figure 1 When the exhaust valve is stuck and cannot move, the air in the pump cannot be discharged, which will cause the standby oil pump to malfunction, affecting the lubricating oil pressure of the small machine, and in severe cases causing the unit to stop operating. Because the one-way valve is inside the main oil tank of the small machine, the unit cannot be disassembled and inspected while it is running and needs to be shut down for inspection. Therefore, the present invention proposes to install an external detachable one-way valve system at the oil pump outlet of the small machine tank to solve the problems existing in the prior art. Summary of the Invention

[0004] In response to the above problems, the present invention proposes to install an externally mounted detachable one-way valve system at the outlet of the oil pump of the small engine oil tank. The externally mounted detachable one-way valve system at the outlet of the oil pump of the small engine oil tank ensures that if the one-way valve becomes stuck, the one-way valve can be inspected and cleaned without affecting the safe operation of the unit, thereby ensuring the normal operation of each oil pump, greatly improving the maintenance efficiency of personnel and ensuring the safe operation of the unit.

[0005] To achieve the purpose of the present invention, the present invention is implemented through the following technical solutions: an external detachable one-way valve system is installed at the outlet of the oil pump of the small engine tank, including a structural component and an identification system. The structural component includes a small engine main oil tank and an oil pump group arranged inside the small engine main oil tank. The oil pump group includes a small engine main oil pump and a small engine DC oil pump, and the output ends of the small engine main oil pump and the small engine DC oil pump are both connected to the oil supply through pipelines. The outside of the small engine main oil tank is connected to a one-way valve, and multiple one-way valves are provided. The one-way valve is connected to the pipeline through a through pipe, and the through pipe is provided with a shut-off valve, and the one-way valve is provided with a microphone;

[0006] The recognition system includes a sound collection module, a data verification module and an analysis and prediction module. The sound collection module is used to collect and store the sound recognized by the microphone. The data verification module is used to perform big data analysis on the recognized sound to determine the current working status of the one-way valve. The analysis and prediction module is used to perform variable analysis on the sound stored in a specified time period and calculate and predict the subsequent problem trends of the one-way valve.

[0007] A further improvement is that in the main oil tank of the small machine, there are at least two groups of main oil pumps for the small machine, at least one group of DC oil pumps for the small machine, and three groups of one-way valves. The pipes at the output ends of the two groups of main oil pumps for the small machine and one group of DC oil pumps for the small machine are respectively connected to the through pipes of the three groups of guide valves.

[0008] A further improvement is that check valves are provided on the pipelines at the output ends of the two groups of small machine main oil pumps and one group of small machine DC oil pumps, and the check valves are provided at the rear end position of the connection between the through pipe and the pipeline.

[0009] A further improvement is that the sound collection module includes a collection unit and a storage unit. The collection unit is connected to a microphone and is used to collect the sound of the one-way valve. The storage unit is used to store the audio data collected by the collection unit and add a timestamp. The storage unit is also used to provide a timestamp query function.

[0010] Further improvements are as follows: the data verification module includes a data packet, a big data learning module and a conclusion output module, the data packet is used to store all standard audio parameter data about the normal sound of the one-way valve, different abnormal sounds and audio decibel data about the abnormal process, and construct a comparison main model based on the standard audio data, and construct a comparison sub-model based on the main model based on the audio decibel parameter data about the abnormal process.

[0011] Further improvements are as follows: the big data learning module has a built-in deep learning algorithm, which is used to input the currently collected one-way valve sound data into the data packet for one-by-one comparison and demonstration. First, the frequency and timbre of the current sound are judged based on the comparison main model to determine whether the current one-way valve sound is normal or abnormal. When it is determined that the sound is abnormal, the decibel and interval duration of the current sound are judged based on the comparison sub-model to determine the process to which the current abnormality belongs. Each stage is compared and demonstrated more than three times.

[0012] Further improvements are as follows: the conclusion output module is used for wired access to the human-computer interaction display in the operator's main control room and for wireless access to the operator's mobile phone APP, and is used to merge the results of more than three comparisons and demonstrations into a data report, and output it to the human-computer interaction display and mobile phone APP. When the result of the comparison and demonstration indicates that the one-way valve is abnormal, an alarm is issued through the human-computer interaction display and mobile phone APP, and the alarm is simultaneously output to the manufacturer's data network according to the current abnormality type, and a processing solution is retrieved online in real time.

[0013] Further improvements are: the analysis and prediction module includes a timing collection module, a data planning module and a fault prediction module. The timing collection module is used to manually set the collection time period, collect the data stored in the storage unit within the specified time period interval according to the manually set standards, and merge them into a prediction unit basis.

[0014] A further improvement is that the data planning module is used to split the prediction unit basis, splitting the various parameters of the audio data into specific vector data, and constructing a line graph based on the fluctuation of each vector data. The fault prediction module is used to analyze the line graphs of different vector data, first analyzing the data change trend, then analyzing the fluctuation amplitude of the data at different time points, and then calculating the time required to reach the abnormal threshold represented by the data based on the change trend and fluctuation amplitude:

[0015] T=(S0-S L )÷N

[0016] Where T refers to the time required to reach abnormality;

[0017] S0 refers to the abnormal threshold value represented by the data;

[0018] SL refers to the value of the data currently collected;

[0019] N refers to the current fluctuation range of the data.

[0020] A further improvement is that the recognition system also includes a log module, which is used to store all data from the sound collection module, data verification module and analysis and estimation module, and synthesize them into text documents, and simultaneously provide keywords, timestamps and data retrieval functions of different modules.

[0021] The beneficial effects of the present invention are:

[0022] 1. The present invention arranges the one-way valve outside the main oil tank of the small machine, and connects it with the pipelines of the main oil pump of the small machine and the DC oil pump of the small machine through a through pipe. When the one-way valve has problems, the through pipe can be disconnected by the cut-off valve to isolate the one-way valve, which is convenient for online replacement after isolation. When the one-way valve becomes stuck, the one-way valve can be checked and cleaned without affecting the safe operation of the unit, thereby ensuring the normal operation of each oil pump, greatly improving the maintenance efficiency of personnel and ensuring the safe operation of the unit.

[0023] 2. The present invention collects the operating sound of the one-way valve through a microphone. Since abnormal sounds will occur at the valve body and the pump body when the one-way valve is abnormal, the sound is collected and compared with the audio data in the data packet through a deep learning algorithm to determine whether the one-way valve is normal or abnormal. Then, based on the detailed parameters of the audio data, when the abnormality is determined, the process to which the abnormality belongs can be determined. Compared with the self-inspection of the oil pump group, the present invention is more timely through audio analysis, and improves the diagnosis of the one-way valve.

[0024] 3. The present invention collects data stored in a storage unit within a specified time period according to manually set standards, splits various parameters of the data into specific vector data, analyzes them using a line graph, determines the change trend and fluctuation amplitude, and thereby calculates the time required to reach the abnormal threshold represented by the data, so as to prevent problems before they occur, identify possible abnormal trends in advance, and reduce risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic diagram of the prior art;

[0026] Figure 2 It is a schematic diagram of the structural components of the present invention;

[0027] Figure 3 Schematic diagram of the identification system of the present invention. DETAILED DESCRIPTION

[0028] In order to deepen the understanding of the present invention, the present invention will be further described in detail below with reference to the examples. The examples are only used to explain the present invention and do not constitute a limitation on the scope of protection of the present invention.

[0029] Example 1

[0030] according to Figure 2 、 3As shown, this embodiment proposes an external detachable one-way valve system installed at the outlet of the oil pump of the small engine tank, including a structural component and an identification system. The structural component includes a small engine main oil tank and an oil pump group arranged inside the small engine main oil tank. The oil pump group includes a small engine main oil pump and a small engine DC oil pump, and the output ends of the small engine main oil pump and the small engine DC oil pump are connected to the oil supply through pipelines. The outside of the small engine main oil tank is connected to a one-way valve, and there are multiple groups of one-way valves. The one-way valve is connected to the pipeline through a through pipe. The one-way valve is The flange is detachable, and a shutoff valve is provided on the through pipe, and a microphone is provided on the one-way valve. When in use, the one-way valve is set outside the main oil tank of the small machine, and is connected to the pipelines of the main oil pump of the small machine and the DC oil pump of the small machine through the through pipe. When the one-way valve has problems, the through pipe can be disconnected through the shutoff valve to isolate the one-way valve, which is convenient for online replacement after isolation. When the one-way valve becomes stuck, the one-way valve can be checked and cleaned without affecting the safe operation of the unit, thereby ensuring the normal operation of each oil pump.

[0031] The recognition system includes a sound collection module, a data verification module, and an analysis and prediction module. The sound collection module is used to collect and store the sound recognized by the microphone. The data verification module is used to perform big data analysis on the recognized sound to determine the current working status of the one-way valve. The analysis and prediction module is used to perform variable analysis on the sound stored in a specified time period to calculate and predict the trend of subsequent problems with the one-way valve. When in use, the operating sound of the one-way valve is collected through the microphone. Since abnormal sounds will occur at the valve body and the pump body when the one-way valve is abnormal, the sound collected can be used to determine whether the one-way valve is normal or abnormal. When the abnormality is determined, the process to which the abnormality belongs can be determined. Compared with the self-inspection of the oil pump group, audio analysis is more timely and improves the diagnosis of the one-way valve.

[0032] In the main oil tank of the small machine, there are two groups of main oil pumps for the small machine, one group of direct current oil pumps for the small machine, and three groups of one-way valves. The pipes at the output ends of the two groups of main oil pumps for the small machine and one group of direct current oil pumps for the small machine are respectively connected to the through pipes of the three groups of guide valves. Check valves are provided on the pipes at the output ends of the two groups of main oil pumps for the small machine and one group of direct current oil pumps for the small machine, and the check valves are provided at the rear end of the connection between the through pipe and the pipe to prevent backflow. The one-way valve is provided outside the main oil tank of the small machine, and is connected to the pipes of the main oil pump for the small machine and the direct current oil pump for the small machine through the through pipe. When a problem occurs with the one-way valve, the through pipe can be disconnected by the shut-off valve to isolate the one-way valve. After isolation, it is convenient to replace the one-way valve online, ensuring that when the one-way valve becomes stuck, the one-way valve can be inspected and cleaned without affecting the safe operation of the unit.

[0033] Example 2

[0034] according to Figure 2 、 3As shown, this embodiment proposes an external detachable one-way valve system installed at the outlet of the oil pump of the small engine tank, including a structural component and an identification system. The structural component includes a small engine main oil tank and an oil pump group arranged inside the small engine main oil tank. The oil pump group includes a small engine main oil pump and a small engine DC oil pump, and the output ends of the small engine main oil pump and the small engine DC oil pump are connected to the oil supply through pipelines. The outside of the small engine main oil tank is connected to a one-way valve, and there are multiple groups of one-way valves. The one-way valve is connected to the pipeline through a through pipe. The one-way valve is The flange is detachable, and a shutoff valve is provided on the through pipe, and a microphone is provided on the one-way valve. When in use, the one-way valve is set outside the main oil tank of the small machine, and is connected to the pipelines of the main oil pump of the small machine and the DC oil pump of the small machine through the through pipe. When the one-way valve has problems, the through pipe can be disconnected through the shutoff valve to isolate the one-way valve, which is convenient for online replacement after isolation. When the one-way valve becomes stuck, the one-way valve can be checked and cleaned without affecting the safe operation of the unit, thereby ensuring the normal operation of each oil pump.

[0035] The recognition system includes a sound collection module, a data verification module, and an analysis and prediction module. The sound collection module is used to collect and store the sound recognized by the microphone. The data verification module is used to perform big data analysis on the recognized sound to determine the current working status of the one-way valve. The analysis and prediction module is used to perform variable analysis on the sound stored in a specified time period to calculate and predict the trend of subsequent problems with the one-way valve. When in use, the operating sound of the one-way valve is collected through the microphone. Since abnormal sounds will occur at the valve body and the pump body when the one-way valve is abnormal, the sound collected can be used to determine whether the one-way valve is normal or abnormal. When the abnormality is determined, the process to which the abnormality belongs can be determined. Compared with the self-inspection of the oil pump group, audio analysis is more timely and improves the diagnosis of the one-way valve.

[0036] The sound collection module includes a collection unit and a storage unit. The collection unit is connected to a microphone and is used to collect the sound of the one-way valve. The storage unit is used to store the audio data collected by the collection unit and add a time stamp. The storage unit is also used to provide a time stamp query function for data storage and data query.

[0037] The data verification module includes a data packet, a big data learning module and a conclusion output module. The data packet is used to store all standard audio parameter data about the normal sound of the one-way valve, different abnormal sounds and audio decibel data about the abnormal process, and to construct a comparison main model based on the standard audio data, and to construct a comparison sub-model based on the main model based on the audio decibel parameter data about the abnormal process.

[0038] The big data learning module has a built-in deep learning algorithm, which is used to input the currently collected one-way valve sound data into a data packet for one-by-one comparison and verification. First, the frequency and timbre of the current sound are judged based on the comparison main model to determine whether the current one-way valve sound is normal or abnormal. When it is determined that the sound is abnormal, the decibel and interval duration of the current sound are judged based on the comparison sub-model to determine the process to which the current abnormality belongs. Each stage is compared and verified more than three times. Specifically, these include: Random Forest, an ensemble learning method that makes predictions by building multiple decision trees and outputting the categories of their classification patterns (for classification tasks); Gradient Boosting, an iterative decision tree algorithm, used for classification, regression, and other learning tasks. It builds a series of weak prediction models (such as decision trees) and combines them into a strong prediction model, attempting to correct the prediction errors of the previous model at each step; Support Vector Machines, which can be used to identify different sound patterns after extracting features from sound data; Neural Networks, particularly Convolutional Neural Networks (CNNs) or Recurrent Neural Networks (RNNs) and their variants (such as LSTMs and GRUs), which are particularly effective for processing time series data and audio signals. Neural Networks can automatically learn features from raw sound data and capture complex patterns; K-Nearest Neighbors, which can directly find the nearest neighbors in the sound feature space and make predictions based on the categories of these neighbors. Implementation Steps: Data Collection and Preprocessing: Obtain a large amount of sound data on the operating status of the check valve. Preprocess the sound data, including denoising and feature extraction (such as Mel-Frequency Cepstral Coefficients (MFCCs) and spectral features). Construct training and test sets: Split the preprocessed data into training and test sets. Ensure that each data point is correctly labeled (i.e., the operating status of the check valve). Select and train the model: Select a machine learning algorithm. Train the model using the training set data. Adjust model parameters to optimize performance. Model evaluation and optimization: Evaluate model performance using the test set. Adjust the model as needed.

[0039] The conclusion output module is connected to the operator's main control room's human-machine interactive display and the operator's mobile phone app via a wired connection. It combines the results of three or more comparisons and demonstrations into a data report, which is then output to the human-machine interactive display and mobile app. If the comparison and demonstration results indicate a one-way valve anomaly, an alarm is issued via the human-machine interactive display and mobile app. This alarm is then simultaneously output to the manufacturer's data network based on the current anomaly type, enabling real-time online retrieval of treatment solutions. Conclusion reports and warnings are output to the human-machine interactive display and mobile app, ensuring that operators have access to data.

[0040] The analysis and prediction module includes a timing collection module, a data planning module and a fault prediction module. The timing collection module is used to manually set the collection time period, collect the data stored in the storage unit within the specified time period according to the manually set standards, and merge them into a prediction unit basis.

[0041] The data planning module is used to split the prediction unit basis, split the various parameters of the audio data into specific vector data, and construct a line graph based on the fluctuation of each vector data. The fault prediction module is used to analyze the line graphs of different vector data, first analyzing the change trend of the data, then analyzing the fluctuation range of the data at different time points, and then calculating the time required to reach the abnormal threshold represented by the data based on the change trend and fluctuation range: using statistical methods (such as linear regression, time series analysis, etc.) to analyze the change trend of the data, based on the analysis results, establish a mathematical model of the data trend, predict the data trend in the future, calculate the fluctuation range of the data at different time points, and analyze whether the fluctuation has periodicity, seasonality or other identifiable patterns by calculating the standard deviation, range or other statistics:

[0042] T=(S0-S L )÷N

[0043] Where T refers to the time required to reach abnormality;

[0044] S0 refers to the abnormal threshold value represented by the data;

[0045] SL refers to the value of the data currently collected;

[0046] N refers to the current fluctuation range of the data.

[0047] The recognition system also includes a log module, which stores all data from the sound acquisition module, data verification module, and analysis and estimation module, and synthesizes it into a text document. It also provides keywords, timestamps, and data retrieval functions for different modules, facilitating human-computer interaction and manual query.

[0048] The system incorporates an externally mounted, removable check valve at the outlet of the oil pump in the oil tank of the small engine. The check valve is located outside the main oil tank of the small engine and is connected to the pipelines of the main oil pump and the direct current oil pump of the small engine via a through pipe. In the event of a problem with the check valve, the through pipe can be disconnected via a shutoff valve, isolating the check valve. This isolation facilitates online replacement, ensuring that if the check valve becomes stuck, the check valve can be inspected and cleaned without affecting the safe operation of the unit, ensuring the normal operation of each oil pump, greatly improving maintenance efficiency, and ensuring the safe operation of the unit. Furthermore, the present invention uses a microphone to collect the operating sound of the check valve. Since abnormalities in the check valve cause abnormal sounds at the valve body and pump body, the sound is collected and compared with the audio data in the data packet through a deep learning algorithm to determine whether the check valve is normal or abnormal. Then, based on the detailed parameters of the audio data, the process to which the abnormality belongs can be determined. Compared with the self-test of the oil pump unit, the present invention is more timely through audio analysis, improving the diagnosis of the check valve. At the same time, the present invention collects data stored in a storage unit within a specified time period according to manually set standards, splits various parameters of the data into specific vector data, performs line graph analysis, determines the change trend and fluctuation amplitude, and thereby calculates the time required to reach the abnormal threshold represented by the data, so as to prevent problems before they occur, identify possible abnormal trends in advance, and reduce risks.

[0049] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the foregoing embodiments. The foregoing embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. An external detachable one-way valve system is installed at the outlet of the oil pump of the small engine oil tank, including structural components and an identification system, characterized by: The structural assembly includes a small engine main oil tank and an oil pump group arranged inside the small engine main oil tank, the oil pump group includes a small engine main oil pump and a small engine DC oil pump, and the output ends of the small engine main oil pump and the small engine DC oil pump are connected to the oil supply through pipelines, the outside of the small engine main oil tank is connected to a one-way valve, and multiple one-way valves are provided. The one-way valve is connected to the pipeline through a through pipe, and the through pipe is provided with a shut-off valve, and the one-way valve is provided with a microphone; The recognition system includes a sound collection module, a data verification module, and an analysis and prediction module. The sound collection module is used to collect and store the sound recognized by the microphone. The data verification module is used to perform big data analysis on the recognized sound to determine the current working status of the one-way valve. The analysis and prediction module is used to perform variable analysis on the sound stored in a specified time period to calculate and predict the trend of subsequent problems with the one-way valve. The sound collection module includes a collection unit and a storage unit. The collection unit is connected to a microphone and is used to collect the sound of the one-way valve. The storage unit is used to store the audio data collected by the collection unit and add a time stamp. The storage unit is also used to provide a time stamp query function. The data verification module includes a data packet, a big data learning module, and a conclusion output module. The data packet is used to store all standard audio parameter data related to normal sounds and different abnormal sounds of the one-way valve and audio decibel data related to abnormal processes, and to construct a comparison main model based on the standard audio data, and to construct a comparison sub-model based on the main model based on the audio decibel parameter data related to abnormal processes; The big data learning module has a built-in deep learning algorithm for inputting the currently collected one-way valve sound data into the data packet and performing a one-by-one comparison and demonstration. First, the frequency and timbre of the current sound are compared and judged based on the main comparison model to determine whether the current one-way valve sound is normal or abnormal. If the sound is determined to be abnormal, the decibel level and interval duration of the current sound are compared and judged based on the sub-comparison model to determine the process to which the current abnormality belongs. Each stage is compared and demonstrated at least three times. The conclusion output module is used for wired access to the human-machine interactive display in the operator's main control room and wireless access to the operator's mobile phone APP, and is used to combine the results of more than three comparisons and demonstrations into a data report and output it to the human-machine interactive display and mobile phone APP. When the result of the comparison and demonstration indicates that the one-way valve is abnormal, an alarm is issued through the human-machine interactive display and mobile phone APP, and the alarm is simultaneously output to the manufacturer's data network according to the current abnormality type, so that a treatment plan can be retrieved online in real time; The analysis and prediction module includes a timing collection module, a data planning module and a fault prediction module. The timing collection module is used to manually set the collection time period, collect the data stored in the storage unit within the specified time period according to the manually set standard, and merge them into a prediction unit basis; The data planning module is used to split the prediction unit basis, splitting the various parameters of the audio data into specific vector data, and constructing a line graph based on the fluctuation of each vector data. The fault prediction module is used to analyze the line graphs of different vector data, first analyzing the change trend of the data, then analyzing the fluctuation amplitude of the data at different time points, and then calculating the time required to reach the abnormal threshold represented by the data based on the change trend and fluctuation amplitude: T=(S0-SL)÷N Where T refers to the time required to reach abnormality; S0 refers to the abnormal threshold value represented by the data; SL refers to the value of the currently collected data; N refers to the current fluctuation range of the data.

2. The system for installing an external detachable one-way valve at the outlet of the oil pump of a small engine oil tank according to claim 1, characterized in that: In the main oil tank of the small machine, there are at least two groups of main oil pumps for the small machine, at least one group of DC oil pumps for the small machine, and three groups of one-way valves. The pipes at the output ends of the two groups of main oil pumps for the small machine and one group of DC oil pumps for the small machine are respectively connected to the through pipes of the three groups of one-way valves.

3. The system of installing an external detachable one-way valve at the outlet of the oil pump of a small engine oil tank according to claim 2, characterized in that: The pipelines at the output ends of the two groups of small machine main oil pumps and one group of small machine DC oil pumps are both provided with check valves, and the check valves are arranged at the rear end position of the connection point between the through pipe and the pipeline.

4. The system for adding an externally mounted detachable one-way valve to the outlet of a small engine oil tank pump according to any one of claims 1 to 3, characterized in that: The recognition system also includes a log module, which is used to store all data from the sound collection module, data verification module and analysis and estimation module, and synthesize them into text documents, and simultaneously provide keywords, timestamps and data retrieval functions for different modules.

Images