Siphon pipeline fault diagnosis alarm system based on big data intelligent learning

Abstract

The invention discloses a siphon pipeline fault diagnosis alarm system based on big data intelligent learning, and relates to the technical field of hydraulic engineering, the siphon pipeline fault diagnosis alarm system comprises a siphon pipeline exhaust device, the siphon pipeline exhaust device comprises a water storage mechanism, a water injection mechanism and a communication mechanism, the top of the water storage mechanism is connected with the water injection mechanism, and the top of the water injection mechanism is connected with the communication mechanism; the bottom of the water storage mechanism is connected with the communication mechanism, the water storage mechanism is further provided with a data acquisition mechanism and a control mechanism, the data acquisition mechanism is a vacuum pressure feedback device, the vacuum pressure feedback device is arranged on the top face of the water storage tank in a penetrating mode, and the control mechanism comprises a controller and a remote controller. According to the siphon pipeline fault diagnosis and alarm system based on big data intelligent learning, prediction of pressure change in the device is realized through big data learning while the overflow capacity in the siphon is not influenced, and when the pressure change exceeds a threshold value, automatic control of a switch valve is completed, so that the operation efficiency of the system is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of water conservancy engineering, and in particular to a siphon pipeline fault diagnosis and alarm system based on big data intelligent learning. Background Art

[0002] The siphon principle is widely used in water conservancy projects. It uses the gravity and velocity difference of the fluid to form negative pressure under certain conditions, thus realizing the automatic attraction and transportation of the fluid. In water conservancy projects, the application of the siphon principle can greatly improve the efficiency of the project and save energy consumption. Through reasonable design and application, more innovations and optimizations of water conservancy projects can be achieved.

[0003] After the siphon is formed, during the water delivery process, the gas dissolved in the water is released due to the change in water pressure in the pipe; in addition, the air mixed in the water well enters the siphon. The gas gathers at the top of the siphon to form an air bag, which reduces the water flow section and reduces the water delivery capacity. To solve the impact of the air bag generated in the siphon on water delivery, the traditional method is to re-evacuate or close the inlet and outlet valves, and inject water into the pipe to stop water delivery.

[0004] Therefore, a siphon pipe fault diagnosis and alarm system based on big data intelligent learning is proposed to solve the above problems. Summary of the invention

[0005] The main purpose of the present invention is to provide a siphon pipe fault diagnosis and alarm system based on big data intelligent learning to solve the problems raised in the above background.

[0006] To achieve the above-mentioned purpose, the technical solution adopted by the present invention is: a siphon pipe fault diagnosis and alarm system based on big data intelligent learning, including a siphon pipe exhaust device, the siphon pipe exhaust device includes a water storage mechanism, a water injection mechanism and a connecting mechanism, the top of the water storage mechanism is connected to the water injection mechanism, the bottom of the water storage mechanism is connected to the connecting mechanism, and the water storage mechanism is also provided with a data acquisition mechanism and a control mechanism.

[0007] Preferably, the water storage mechanism is a hollow water storage tank, and the water injection mechanism includes a water injection pipe, a water injection gate valve and a water injection funnel. A first opening is provided on the top surface of the water storage tank, a water injection pipe is provided in the first opening, a first flange is provided at the end of the water injection pipe away from the water storage tank, a second flange is provided on the top surface of the first flange, the first flange and the second flange are connected by bolts, and a water injection gate valve is provided between the first flange and the second flange.

[0008] Preferably, the first flange and the second flange have the same structural size and are provided with a circular hole passing through the middle position, the water injection pipe is provided in the circular hole of the first flange, and a water injection funnel is provided on the top surface of the second flange, and one end of the water injection funnel is provided in the circular hole of the second flange.

[0009] Preferably, the connecting mechanism includes an air intake gate valve and a connecting pipe, a second opening is provided through the bottom surface of the water storage tank, a third flange is provided at the second opening, a fourth flange is provided on the bottom surface of the third flange, the third flange and the fourth flange are connected by bolts, and an air intake gate valve is provided between the third flange and the fourth flange.

[0010] Preferably, the third flange and the fourth flange have the same structural size and are provided with a circular hole through the middle position, the connecting pipe is arranged in the circular hole of the fourth flange, and the end of the connecting pipe away from the water storage tank is arranged in the siphon.

[0011] Preferably, the data acquisition mechanism is a vacuum pressure feedback device, and the vacuum pressure feedback device is arranged on the top surface of the water storage tank. The control mechanism includes a controller and a remote controller, and the controller is electrically connected to the vacuum pressure feedback device, the air intake gate valve, and the water injection gate valve respectively, and the controller is bidirectionally data connected to the remote controller.

[0012] Preferably, the following process is performed between the remote controller and the controller:

[0013] S1: The controller sends the pressure information collected by the vacuum pressure feedback device to the remote controller;

[0014] S2: The remote controller performs analysis and calculation based on the collected pressure information, and feeds back the analysis and calculation results to the controller;

[0015] S3: The controller controls the opening and closing of the air inlet valve and the water injection valve according to the analysis and calculation results.

[0016] Preferably, the step S1 is specifically:

[0017] The vacuum pressure feedback device includes a pressure sensor and a signal converter. The pressure sensor is data-connected to the signal converter. The pressure sensor collects the pressure value in the water storage tank at intervals and sends the collected pressure value to the signal converter. The signal converter converts the received pressure value information into an electrical signal, and finally sends the converted electrical signal to the remote controller.

[0018] Preferably, the step S2 specifically comprises the following steps:

[0019] S21: storing the pressure values ​​in the water storage tank collected at intervals into a set Y, and recording the pressure value in the water storage tank each time the air inlet valve is closed and the water injection valve is opened, and storing the pressure value in a set Z;

[0020] S22: Draw a first pressure value change curve according to the data in set Y, draw a second pressure value change curve according to the data in set Z, create a pressure value learning model according to the second pressure value change curve, and obtain a threshold range of the pressure value in the water storage tank when the air intake valve and the water injection valve are opened according to the pressure value learning model;

[0021] The pressure value neural network model is created, and the threshold interval of the pressure value in the water storage tank when the air inlet valve and the water injection valve are opened is obtained according to the pressure value neural network model, including:

[0022] The maximum value, minimum value, average value and standard deviation are extracted from the second pressure value variation curve as features for model input, wherein the minimum value is calculated as:

[0023] x min =min(x1, x2, ..., x n );

[0024] The maximum value is calculated as:

[0025] x max =max(x1, x2, ..., x n );

[0026] The average is calculated as:

[0027]

[0028] The standard deviation is calculated as:

[0029]

[0030] The established neural network model is trained using the historical data in the collected set Z;

[0031] Use the validation dataset to evaluate the performance of the model and tune it as needed to get the stress value threshold.

[0032] Preferably, the step S3 is specifically:

[0033] According to the pressure value threshold, the pressure value when the air intake gate valve is opened and the water injection gate valve is closed is determined; and the pressure value when the air intake gate valve is closed and the water injection gate valve is opened is determined.

[0034] The present invention has the following beneficial effects:

[0035] 1. In the present invention, by providing an exhaust device for the siphon pipe, the effect of the air bag on water delivery is reduced without stopping the water supply and reducing the water delivery capacity.

[0036] 2. In the present invention, by adopting a neural network learning model based on big data, the autonomous learning ability of the device is effectively improved. By collecting the pressure values ​​in the device and using historical pressure values ​​to train the model, the vacuum degree and vacuum degree change law in the water storage tank can be effectively monitored, fault self-diagnosis and remote alarm can be achieved, and the equipment can be repaired immediately. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a schematic diagram of the overall system architecture of a siphon pipeline fault diagnosis and alarm system based on big data intelligent learning according to the present invention;

[0038] Figure 2 A schematic diagram of a siphon pipe exhaust device of a siphon pipe fault diagnosis and alarm system based on big data intelligent learning of the present invention;

[0039] Figure 3 This is a flowchart of execution between a remote controller and a controller of a siphon pipeline fault diagnosis and alarm system based on big data intelligent learning of the present invention;

[0040] Figure 4 This is a flowchart of the execution of a remote controller of a siphon pipeline fault diagnosis and alarm system based on big data intelligent learning in the present invention.

[0041] Description of Figure Numbers:

[0042] 1. Flange; 2. Air inlet gate valve; 3. Water storage tank; 4. Water filling funnel; 5. Vacuum pressure feedback device; 6. Water filling gate valve. DETAILED DESCRIPTION

[0043] The present invention will be further described below in conjunction with specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention. In addition, it should be understood that after reading the content taught by the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms fall within the scope limited by the application equally.

[0044] In the present invention, terms such as "upper", "lower", "left", "right", "front", "back", "vertical", "horizontal", "side", "bottom", etc. indicate directions or positional relationships based on the directions or positional relationships shown in the accompanying drawings. They are relational words determined only for the convenience of describing the structural relationships of the various parts or elements of the present invention, and do not specifically refer to any part or element in the present invention and should not be understood as limitations on the present invention.

[0045] In the present invention, terms such as "fixed connection", "connected", "connection", etc. should be understood in a broad sense, indicating that it can be fixedly connected, integrally connected or detachably connected; it can be directly connected or indirectly connected through an intermediate medium. Relevant scientific research or technical personnel in this field can determine the specific meanings of the above terms in the present invention according to specific circumstances, and they should not be understood as limiting the present invention.

[0046] Example:

[0047] like Figures 1 to 4 As shown, a siphon pipe fault diagnosis and alarm system based on big data intelligent learning includes a siphon pipe exhaust device, which includes a water storage mechanism, a water injection mechanism and a connecting mechanism. The top of the water storage mechanism is connected to the water injection mechanism, and the bottom of the water storage mechanism is connected to the connecting mechanism. A data acquisition mechanism and a control mechanism are also provided on the water storage mechanism.

[0048] The water storage mechanism is a hollow water storage tank, and the water injection mechanism includes a water injection pipe, a water injection gate valve and a water injection funnel. A first opening is set on the top surface of the water storage tank, and a water injection pipe is set in the first opening. A first flange is set on the end of the water injection pipe away from the water storage tank, and a second flange is set on the top surface of the first flange. The first flange and the second flange are connected by bolts, and a water injection gate valve is set between the first flange and the second flange.

[0049] The first flange and the second flange have the same structure and size, and are provided with a circular hole in the middle. The water injection pipe is provided in the circular hole of the first flange, and a water injection funnel is provided on the top surface of the second flange, and one end of the water injection funnel is provided in the circular hole of the second flange.

[0050] The connecting mechanism includes an air intake gate valve and a connecting pipe. A second opening is provided on the bottom surface of the water storage tank, a third flange is provided at the second opening, a fourth flange is provided on the bottom surface of the third flange, the third flange and the fourth flange are connected by bolts, and an air intake gate valve is provided between the third flange and the fourth flange.

[0051] The third flange and the fourth flange have the same structure and size, and are provided with a circular hole in the middle. The connecting pipe is arranged in the circular hole of the fourth flange, and the end of the connecting pipe away from the water storage tank is arranged in the siphon.

[0052] The data acquisition mechanism is a vacuum pressure feedback device, which is arranged on the top surface of the water storage tank. The control mechanism includes a controller and a remote controller. The controller is electrically connected to the vacuum pressure feedback device, the air intake gate valve, and the water injection gate valve respectively, and the controller and the remote controller are bidirectionally data connected.

[0053] The exhaust device is set at the highest point of the siphon pipe. The controller collects data from the vacuum pressure feedback device and controls the opening and closing of the air inlet gate valve and the water injection gate valve. The data collected by the controller is transmitted to 8 remote controllers in real time through the Internet of Things. The remote controller can remotely control the opening and closing of the air inlet gate valve and the water injection gate valve. During normal operation, the siphon pipe and the water storage tank are vacuum, the air inlet gate valve is opened, and the water injection gate valve is closed; the gas mixed in the pipe and the gas dissolved in the water enter the water storage tank through the air inlet gate valve at the high point of the siphon pipe;

[0054] The gas does not occupy the water transmission section and does not affect the flow capacity.

[0055] As air gradually accumulates in the water tank during the water delivery process, the vacuum in the water tank gradually decreases. When the feedback data from the vacuum pressure feedback device reaches a certain value, the controller closes the air inlet valve, opens the water injection valve to fill the water delivery pipe with water, then closes the water injection valve and opens the air inlet valve to complete the exhaust.

[0056] The following process is performed between the remote controller and the controller:

[0057] S1: The controller sends the pressure information collected by the vacuum pressure feedback device to the remote controller;

[0058] S2: The remote controller performs analysis and calculation based on the collected pressure information, and feeds back the analysis and calculation results to the controller;

[0059] S3: The controller controls the opening and closing of the air inlet valve and the water injection valve according to the analysis and calculation results.

[0060] Step S1 is specifically:

[0061] The vacuum pressure feedback device includes a pressure sensor and a signal converter. The pressure sensor is data-connected to the signal converter. The pressure sensor collects the pressure value in the water storage tank at intervals and sends the collected pressure value to the signal converter. The signal converter converts the received pressure value information into an electrical signal, and finally sends the converted electrical signal to the remote controller.

[0062] Step S2 specifically includes the following steps:

[0063] S21: storing the pressure values ​​in the water storage tank collected at intervals into a set Y, and recording the pressure value in the water storage tank each time the air inlet valve is closed and the water injection valve is opened, and storing the pressure value in a set Z;

[0064] S22: Draw a first pressure value change curve according to the data in set Y, draw a second pressure value change curve according to the data in set Z, create a pressure value learning model according to the second pressure value change curve, and obtain a threshold range of the pressure value in the water storage tank when the air intake valve and the water injection valve are opened according to the pressure value learning model;

[0065] Create a pressure value neural network model, and obtain the threshold range of the pressure value in the water storage tank when the air intake valve and the water injection valve are opened according to the pressure value neural network model, including:

[0066] The maximum value, minimum value, average value and standard deviation are extracted from the second pressure value variation curve as features for model input, wherein the minimum value is calculated as:

[0067] x min =min(x1, x2, ..., x n );

[0068] The maximum value is calculated as:

[0069] x max =max(x1, x2, ..., x n );

[0070] The average is calculated as:

[0071]

[0072] The standard deviation is calculated as:

[0073]

[0074] The established neural network model is trained using the historical data in the collected set Z;

[0075] Use the validation dataset to evaluate the performance of the model and tune it as needed to get the stress value threshold.

[0076] Among them, the model is trained as follows:

[0077] 1. Preprocess the data:

[0078] Normalization: Scale the data set to a specific range (usually 0-1) to eliminate the dimensional effects between different features. The calculation formula is:

[0079]

[0080] Standardization: adjust the mean of the data to 0 and the standard deviation to 1. The calculation formula is:

[0081]

[0082] 2. Define the loss function:

[0083] Common loss functions include mean square error (MSE) and cross entropy. For example, the formula for mean square error is:

[0084]

[0085] 3. Back propagation algorithm:

[0086] The gradient of the loss function for each parameter is calculated by the chain rule, and then the parameters are updated. The gradient calculation formula is:

[0087]

[0088] 4. Use gradient descent algorithm for optimization:

[0089]

[0090] 5. Divide the data set into training set, validation set and test set. The training set is used to train the model, the validation set is used to adjust the hyperparameters, and the test set is used to evaluate the model performance;

[0091] 6. Perform iterative training using the training data and the selected optimization algorithm. In each iteration, the predictions are calculated through forward propagation and the weights are updated through back propagation. During training, the performance on the validation set is monitored to prevent overfitting.

[0092] Step S3 is specifically:

[0093] According to the pressure value threshold, the pressure value when the air intake gate valve is opened and the water injection gate valve is closed is determined; and the pressure value when the air intake gate valve is closed and the water injection gate valve is opened is determined.

[0094] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A siphon pipeline fault diagnosis and alarm system based on big data intelligent learning, including a siphon pipeline exhaust device, characterized in that: The siphon pipe exhaust device includes a water storage mechanism, a water injection mechanism and a connecting mechanism. The top of the water storage mechanism is connected to the water injection mechanism, and the bottom of the water storage mechanism is connected to the connecting mechanism. The water storage mechanism is also provided with a data acquisition mechanism and a control mechanism.

2. According to the big data intelligent learning-based siphon pipeline fault diagnosis and alarm system of claim 1, it is characterized in that: The water storage mechanism is a hollow water storage tank, and the water injection mechanism includes a water injection pipe, a water injection gate valve and a water injection funnel. A first opening is provided on the top surface of the water storage tank, and a water injection pipe is provided in the first opening. A first flange is provided at the end of the water injection pipe away from the water storage tank, and a second flange is provided on the top surface of the first flange. The first flange and the second flange are connected by bolts, and a water injection gate valve is provided between the first flange and the second flange.

3. A siphon pipeline fault diagnosis and alarm system based on big data intelligent learning according to claim 2, characterized in that: The first flange and the second flange have the same structure and size, and are provided with a circular hole in the middle. The water injection pipe is provided in the circular hole of the first flange, and a water injection funnel is provided on the top surface of the second flange, and one end of the water injection funnel is provided in the circular hole of the second flange.

4. The siphon pipeline fault diagnosis and alarm system based on big data intelligent learning according to claim 2 is characterized in that: The connecting mechanism includes an air intake gate valve and a connecting pipe. A second opening is provided on the bottom surface of the water storage tank, a third flange is provided at the second opening, a fourth flange is provided on the bottom surface of the third flange, the third flange and the fourth flange are connected by bolts, and an air intake gate valve is provided between the third flange and the fourth flange.

5. A siphon pipeline fault diagnosis and alarm system based on big data intelligent learning according to claim 4, characterized in that: The third flange and the fourth flange have the same structure and size, and are provided with a circular hole in the middle. The connecting pipe is arranged in the circular hole of the fourth flange, and the end of the connecting pipe away from the water tank is arranged in the siphon.

6. The siphon pipeline fault diagnosis and alarm system based on big data intelligent learning according to claim 2 is characterized in that: The data acquisition mechanism is a vacuum pressure feedback device, which is arranged on the top surface of the water storage tank. The control mechanism includes a controller and a remote controller. The controller is electrically connected to the vacuum pressure feedback device, the air intake gate valve, and the water injection gate valve respectively. The controller is bidirectionally data connected to the remote controller. The controller is specifically a single-chip microcomputer, which is used to collect data from the siphon pipe exhaust device and control the gate valve switch. The remote controller is specifically a remote control platform, which is used to receive and send data instructions.

7. A siphon pipeline fault diagnosis and alarm system based on big data intelligent learning according to claim 6, characterized in that: The following process is performed between the remote controller and the controller: S1: The controller sends the pressure information collected by the vacuum pressure feedback device to the remote controller; S2: The remote controller performs analysis and calculation based on the collected pressure information, and feeds back the analysis and calculation results to the controller; S3: The controller controls the opening and closing of the air inlet valve and the water injection valve according to the analysis and calculation results.

8. The siphon pipeline fault diagnosis and alarm system based on big data intelligent learning according to claim 7 is characterized in that: The step S1 is specifically: The vacuum pressure feedback device includes a pressure sensor and a signal converter. The pressure sensor is data-connected to the signal converter. The pressure sensor collects the pressure value in the water storage tank at intervals and sends the collected pressure value to the signal converter. The signal converter converts the received pressure value information into an electrical signal, and finally sends the converted electrical signal to the remote controller.

9. The siphon pipeline fault diagnosis and alarm system based on big data intelligent learning according to claim 8 is characterized in that: The step S2 specifically comprises the following steps: S21: storing the pressure values ​​in the water storage tank collected at intervals into a set Y, and recording the pressure value in the water storage tank each time the air inlet valve is closed and the water injection valve is opened, and storing the pressure value in a set Z; S22: Draw a first pressure value change curve according to the data in set Y, draw a second pressure value change curve according to the data in set Z, create a pressure value learning model according to the second pressure value change curve, and obtain a threshold range of the pressure value in the water storage tank when the air intake valve and the water injection valve are opened according to the pressure value learning model; The pressure value neural network model is created, and the threshold interval of the pressure value in the water storage tank when the air inlet valve and the water injection valve are opened is obtained according to the pressure value neural network model, including: The maximum value, minimum value, average value and standard deviation are extracted from the second pressure value variation curve as features for model input, wherein the minimum value is calculated as: x min min(x1,x2,...,x n ): The maximum value is calculated as: x max =max(x1,x2,…,x n ); The average is calculated as: The standard deviation is calculated as: The established neural network model is trained using the historical data in the collected set Z; Use the validation dataset to evaluate the performance of the model and tune it as needed to get the stress value threshold.

10. The siphon pipeline fault diagnosis and alarm system based on big data intelligent learning according to claim 9 is characterized in that: The step S3 is specifically: According to the pressure value threshold, the pressure value when the air intake gate valve is opened and the water injection gate valve is closed is determined; and the pressure value when the air intake gate valve is closed and the water injection gate valve is opened is determined.

Images