Heat supply pipeline leakage identification method, system and equipment based on secondary decomposition and BiLSTM and medium

Abstract

The invention provides a heat supply pipeline leakage identification method, system and device based on secondary decomposition and BiLSTM and a medium, and the method comprises the following steps: obtaining an actually measured pressure value of a heat supply pipeline in a preset time period, and taking the obtained actually measured pressure value as input data; carrying out VMD-EMD secondary decomposition on the input data to obtain a residual component; taking the obtained residual component as the input of the BiLSTM network model obtained by construction to obtain a predicted pressure value; the obtained predicted pressure value is compared with a preset fixed early warning threshold value, and whether the heat supply pipeline leaks or not is judged according to a comparison result; according to the invention, the leakage fault can be accurately identified, and the accuracy of fault diagnosis is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of heat supply pipeline leakage identification and detection technology, and relates to a heat supply pipeline leakage identification method, system, device and medium based on quadratic decomposition and BiLSTM. Background Technique

[0002] The heat supply pipe network is a key link connecting the heat source and heat users, and is of great significance to the safe and stable operation of the heat supply system. With the increasing complexity of the pipe network topology and hydraulic conditions, and the increasing service life of pipelines year by year, the failure rate also increases. Common failures include leakage and blockage, among which leakage is the most important problem affecting pipeline safety. Pipeline leakage will not only cause waste of water and heat and serious economic losses, but also affect hygiene, transportation, etc., and even endanger people's lives in severe cases. Therefore, it has important practical significance to quickly and accurately determine the fault type, location and degree of the heat supply pipeline system, ensure the safe operation of the system and improve the operation efficiency.

[0003] Existing pipeline leakage detection methods mainly include acoustic detection method, thermal imaging detection method and negative pressure wave detection method, etc. Due to advantages such as low cost, simple operation, high sensitivity and accurate positioning accuracy, negative pressure wave detection has received extensive attention. Its principle is that when the pipeline is disturbed by noise or there are working condition changes and pipeline leakage, a negative pressure wave will be generated in the pipeline. There are significant differences in the negative pressure wave characteristics between the two. Identifying the differences can identify whether it is a normal working condition or a leakage occurs. And signal denoising is the research focus of negative pressure wave detection. Currently, common denoising methods include wavelet analysis (WT), empirical mode decomposition (EMD), variational mode decomposition (VMD), etc. Among them, wavelet analysis has strong denoising ability, but its denoising effect is greatly affected by the mother function and threshold, and has high empirical requirements for parameter setting. Empirical mode decomposition does not require pre-setting of basis functions, but it is prone to mode mixing and end effect phenomena during the signal decomposition process. The residual term obtained by variational mode decomposition is still relatively complex.

[0004] The deficiency of negative pressure wave detection lies in that it is often interfered by the adjustment working conditions and causes false alarms. With the rich application of data mining algorithms combined with deep learning in the industrial field, its powerful data learning ability has certain adaptability and can solve the deficiencies of negative pressure wave detection to a certain extent. LUO used an improved convolutional neural network (ICNN) for pipeline leakage identification. HAN proposed a flow prediction method based on a convolutional long short-term memory network (CNN-LSTM) to monitor pipeline leakage. MA used an improved AlexNet convolutional network to identify pipeline leakage data of multiple heat stations. Although the above studies used deep learning algorithms for pipeline leakage detection, as a typical one-dimensional time series variable, the data features and time features contained in the negative pressure wave signal have not been deeply mined, and there is still much room for improvement in leakage detection accuracy. And the above methods all output unidirectionally from front to back. When detecting and locating pipeline leakage, the data after leakage is of great significance for detection and location. Therefore, bidirectional mining of data information will help achieve accurate positioning. Summary of the Invention

[0005] The purpose of the present invention is to provide a method, system, device and medium for identifying heat supply pipeline leakage based on quadratic decomposition and BiLSTM, which solves the defects of the existing heat supply pipeline leakage identification methods, such as low detection accuracy and inability to locate pipeline leakage points.

[0006] In order to achieve the above purpose, the technical solution adopted by the present invention is: A method for identifying heat supply pipeline leakage based on quadratic decomposition and BiLSTM provided by the present invention includes the following steps: Obtain the measured pressure values within a preset time period of the heat supply pipeline, and use the obtained measured pressure values as input data; Perform VMD-EMD quadratic decomposition on the input data to obtain residual components; Use the obtained residual components as the input of the constructed BiLSTM network model to obtain predicted pressure values; Compare the obtained predicted pressure values with a preset fixed warning threshold, and judge whether the heat supply pipeline leaks according to the comparison result.

[0007] Preferably, the method for performing VMD-EMD quadratic decomposition on the input data to obtain quadratic decomposition components is as follows: Use VMD to decompose the input data to obtain modal components IMF including residual terms; Use EMD to decompose the residual terms to obtain residual components.

[0008] Preferably, the method for constructing the BiLSTM network model is as follows: Obtain historical pressure signals under normal conditions, leakage conditions, and regulating valve conditions, and use the historical pressure signals as the original input data; Perform VMD-EMD secondary decomposition on the original input data to obtain residual components; Use the obtained residual components to train the constructed BiLSTM network model to obtain the trained BiLSTM network model.

[0009] Preferably, the preset fixed warning threshold is obtained as follows: Use the residual components to train the constructed BiLSTM network model to obtain prediction data; Compare the obtained prediction data with the real data to calculate the prediction error value; Set the fixed warning threshold according to the obtained prediction error value.

[0010] Preferably, compare the obtained predicted pressure value with the preset fixed warning threshold, and judge whether the heat supply pipeline leaks according to the comparison result. The specific method is as follows: Calculate the prediction error between the obtained predicted pressure value and the measured pressure value; Compare the obtained prediction error with the preset fixed warning threshold. Among them, when the prediction error is greater than the preset fixed warning threshold, the heat supply pipeline leaks; otherwise, the heat supply pipeline is normal.

[0011] A heat supply pipeline leakage identification system based on secondary decomposition and BiLSTM includes: A pressure signal acquisition unit for acquiring the measured pressure value of the heat supply pipeline within a preset time period and using the obtained measured pressure value as input data; A data decomposition unit for performing VMD-EMD secondary decomposition on the input data to obtain residual components; A model prediction unit for using the obtained residual components as the input of the constructed BiLSTM network model to obtain the predicted pressure value; A judgment unit for comparing the obtained predicted pressure value with the preset fixed warning threshold and judging whether the heat supply pipeline leaks according to the comparison result.

[0012] An electronic device includes a processor and a memory. A computer instruction is stored on the memory. When the computer instruction is executed by the processor, the electronic device executes the described method.

[0013] A computing device cluster includes at least one computing device, and each computing device includes a processor and a memory; The processor of the at least one computing device is configured to execute instructions stored in the memory of the at least one computing device, so that the computing device cluster executes the method described above.

[0014] A computer program product includes computer-executable instructions thereon, and the computer-executable instructions, when executed, implement the method described above.

[0015] A computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions, when executed by a processor, implement the method described above.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: A method for identifying heat supply pipeline leakage based on quadratic decomposition and BiLSTM provided by the present invention takes the negative pressure wave detection method as the theoretical basis, and collects the negative pressure wave signals of the heat supply pipeline under different working conditions and during leakage; uses the VMD-EMD quadratic decomposition method to denoise the negative pressure wave signals, reduces the mode mixing phenomenon to a certain extent, and reduces the complexity of the original negative pressure wave signals, which is beneficial for neural networks to train and learn; then trains the quadratic decomposition components through a bidirectional long short-term memory neural network system (BiLSTM), identifies and learns the negative pressure wave signals under different working conditions and during leakage, and sets a reasonable prediction threshold according to the prediction error to accurately identify the leakage fault and improve the accuracy of fault diagnosis. During the signal denoising process, a quadratic decomposition method based on VMD-EMD is proposed. By combining the advantages of the two decomposition methods, the non-stationarity of time series with high complexity and strong non-linearity can be reduced, and the problem of mode mixing is solved to a certain extent; during the fault diagnosis process, by bidirectionally mining the temporal information in the data through BiLSTM for modeling, the forward and backward information of the sequence data can be captured simultaneously, better identifying the characteristic differences of the negative pressure wave under different working conditions and during pipeline leakage, realizing the accurate detection and positioning of heat supply pipeline leakage, and providing a direction for the heat supply pipeline leakage identification and detection technology.

[0017] Furthermore, the residual left by the variational mode decomposition (VMD) is decomposed by the empirical mode decomposition (EMD), which can reduce the complexity and non-stationarity of the time series, as well as the mode mixing phenomenon in the decomposition.

[0018] Furthermore, the bidirectional long short-term memory network (BiLSTM) consists of two independently existing unidirectional long short-term memory networks (LSTM) with opposite directions, which can capture the forward and backward information of the sequence data simultaneously, combine the data before and after leakage, better identify the characteristic differences of the negative pressure wave under different working conditions and during pipeline leakage, realize the accurate detection and positioning of heat supply pipeline leakage, and improve the detection and identification accuracy. Description of the Drawings

[0019] Figure 1 is the flowchart of the leakage detection and location process according to an embodiment of the present invention; Figure 2 is the schematic diagram of the VMD-EMD secondary decomposition algorithm according to an embodiment of the present invention. Detailed implementation manners

[0020] In the following description, for the purpose of illustration rather than limitation, specific details such as specific system architectures and technologies are presented in order to provide a thorough understanding of the embodiments of the present application. However, those skilled in the art should understand that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from obscuring the description of the present application.

[0021] It should be understood that when used in the specification of the present application and the appended claims, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0022] It should also be understood that the term "and / or" as used in the specification of the present application and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0023] As used in the specification of the present application and the appended claims, the term "if" can be interpreted as "when", "once", "in response to determining", or "in response to detecting" according to the context. Similarly, the phrase "if determined" or "if [the described condition or event] is detected" can be interpreted as meaning "once determined", "in response to determining", "once [the described condition or event] is detected", or "in response to detecting [the described condition or event]" according to the context.

[0024] In addition, in the description of the specification of the present application and the appended claims, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0025] References to "one embodiment" or "some embodiments" in the description of this application mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of this application. Thus, statements such as "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc. that appear in different places in this specification do not necessarily all refer to the same embodiment, but rather mean "one or more but not all embodiments", unless otherwise specifically emphasized. The terms "comprising", "including", "having" and their variants all mean "including but not limited to", unless otherwise specifically emphasized.

[0026] Embodiment 1 As Figure 1 shown, a heating pipeline leakage identification method based on quadratic decomposition and BiLSTM provided in this embodiment specifically includes the following steps: Step 1, in the offline stage, collect historical pressure signals under normal conditions, leakage conditions, and valve adjustment conditions, and use the historical pressure signals as the original input data; Step 2, perform VMD-EMD quadratic decomposition on the original input data to remove high-frequency noise. The proposed VMD-EMD quadratic decomposition algorithm process is as Figure 2 shown, and the steps are as follows: S21, construct a constrained variational model of VMD, and use the constrained variational model to decompose the original input data:

[0027] In the formula, is the decomposed mode component IMF, is the original input data, is the iterative signal, is the number of decomposition modes. is the impulse function, is the center frequency of each component, represents the partial derivative of the function with respect to t, represents the constraint condition; t is time.

[0028] S22, introduce the Lagrange multiplier and the bandwidth parameter , construct an augmented Lagrangian function, and its expression is as follows, to convert the constrained variational problem into an unconstrained variational problem and obtain an unconstrained optimization problem:

[0029] In the formula, represents the bandwidth parameter; represents the Lagrange multiplier.

[0030] S23. Iteratively update the variables using the multiplicative operator splitting method, , and , to solve the unconstrained optimization problem, where the update formula is:

[0031] the update formula is:

[0032] the update formula is:

[0033] The iteration termination condition is:

[0034] S24. Start the iteration from n = 0, and the number of iterations is taken as , starting from 1 until the preset decomposition mode number , The value is determined by observing whether the center frequency iteration curves overlap. If the center frequency iteration curves overlap, the decomposition mode number takes as the preset decomposition mode number.

[0035] S25. Repeat S23 until the iteration termination condition is satisfied.

[0036] S26. The IMF components obtained by VMD decomposition contain a residual term . Decompose the residual term using EMD, find the local extreme points of the residual term R, and respectively determine the upper and lower envelopes of and through cubic spline interpolation.

[0037] S27. Use the upper and lower envelopes of the residual term , obtained in S26 to calculate the mean value and the initial component :

[0038]

[0039] S28. Judge the initial component Whether the difference between the number of zero-crossing points and extreme points of each IMF is one, and the upper and lower envelopes of local extrema are reciprocal to each other. If the conditions are met, the initial component sought is the first IMF component; if the conditions are not met, repeat S26 and S27 until the initial component meets the conditions.

[0040] S29. Subtract the obtained initial component from to form a new original sequence . Then return to the first step. After repeating n times until no more IMFs can be decomposed, the residual component is obtained:

[0041]

[0042] Step 3. Train each residual component obtained in Step 2 using a BiLSTM network. The BiLSTM performs the following calculation processes respectively.

[0043] S31. First, apply a forget gate to help the BiLSTM decide what information to discard from the cell state. represents the activation function of the forget gate. Among them, the forget gate activation function is as follows:

[0044] Among them, is the output of the forget gate; is the weight matrix of the forget gate; is the hidden state at the previous moment; is the VMD-EMD secondary decomposition component; is the bias term of the forget gate; is the Sigmoid activation function. Through the Sigmoid activation function, a value between 0 and 1 is output. 0 means completely forgetting the last state, and 1 means completely retaining the last state.

[0045] In the BiLSTM, the forget gate discards the information in the cell state. In the forward LSTM, the forget gate decides which information to discard from the forward cell state, and the discarded information is the old information irrelevant to the current task. In the reverse LSTM, the forget gate decides which information to discard from the reverse cell state, and the discarded information is the old information irrelevant to the current task.

[0046] The output of the forget gate is multiplied by the cell state at the previous moment to determine which information is retained or discarded:

[0047] Wherein: is the cell state at the current moment; represents element-wise multiplication; is the output of the input gate; is the candidate cell state.

[0048] S32, The BiLSTM determines which new information will be stored in the new cell state through the input gate. The formula of the input gate layer is as follows:

[0049] S33, The input gate determines the information to be updated. Construct a vector to store the new candidate values to be added to the new cell state, candidate state:

[0050] S34, The old cell state is updated to have an estimated and new cell state . Specifically, the old cell state is multiplied by to forget the information from the previous state. The candidate values are multiplied by the input gate to determine the new amount of information to be updated to the new cell state, update state:

[0051] S35, The output gate filters and outputs the cell state as , output gate:

[0052] In addition, the activation function acts on the new cell state and then multiplies by the output to provide the required information, hidden layer output :

[0053] Finally, the predicted value is obtained by the output layer:

[0054] Step 4, According to Step 3, the predicted data is obtained through training, and compared with the true value to obtain the prediction error. The prediction error calculation formula is:

[0055] In the formula, is the number of samples; is the predicted value; is the measured value.

[0056] Step 5: Design a reasonable fixed warning threshold TV according to the prediction error in Step 4.

[0057] Step 6: In the online stage, take the measured pressure value within the set time as input data, and use the same VMD-EMD quadratic decomposition method to process the input data; then send the data into the trained BiLSTM network to obtain the predicted pressure value; calculate the prediction error FE between the obtained predicted pressure value and the measured pressure value. When the prediction error FE is greater than the warning threshold TV, send a pipeline fault signal and record the fault moment; combine the detection results of multiple pressure sensors to achieve the leakage point positioning.

[0058] This application combines the VMD-EMD quadratic decomposition method and the bidirectional long short-term memory neural network and applies them to the heating pipeline leakage identification and detection technology. By using the VMD-EMD quadratic decomposition method to reduce the complexity of the original negative pressure wave signal, it is beneficial for the bidirectional long short-term memory neural network to train and learn the negative pressure wave signal under different working conditions and leakage conditions; in addition, the bidirectional long short-term memory neural network itself can simultaneously capture the forward and backward information of the sequence data, and better identify the characteristic differences of the negative pressure wave under different working conditions and pipeline leakage. These two points jointly improve the accuracy of heating pipeline leakage detection and positioning from the perspectives of data and model.

[0059] Embodiment 2 A heating pipeline leakage identification system based on quadratic decomposition and BiLSTM provided in this embodiment includes: A pressure signal acquisition unit for acquiring the real-time pressure signal of the heating pipeline and using the obtained pressure signal as input data; A data decomposition unit for performing VMD-EMD quadratic decomposition on the input data to obtain residual components; A model prediction unit for using the obtained residual components as the input of the constructed BiLSTM network model to obtain prediction data; A judgment unit for comparing the obtained prediction data with a preset fixed warning threshold and judging whether the heating pipeline leaks according to the comparison result.

[0060] Embodiment 3 This embodiment also provides a computing device. The computing device includes: a bus, a processor, a memory, and a communication interface. The processor, the memory, and the communication interface communicate through the bus. The computing device can be a server or a terminal device. It should be understood that this application does not limit the number of processors and memories in the computing device.

[0061] The bus can be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus can include a path for transmitting information between various components of the computing device (e.g., memory, processor, communication interface).

[0062] The processor can include any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a Tensor Processing Unit (TPU), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), a microprocessor (MP), or a digital signal processor (DSP), etc.

[0063] The memory can include volatile memory, such as random access memory (RAM). The processor can also include non-volatile memory, such as read-only memory (ROM), flash memory, a hard disk drive (HDD), or a solid state drive (SSD).

[0064] The memory stores executable program code, and the processor executes the executable program code to respectively implement the functions of the foregoing first generation module, second generation module, and adjustment module, thereby implementing, for example, *methods, etc. That is, instructions for the methods and functions of the computing device involved in any of the above embodiments can be stored on the memory.

[0065] The communication interface uses a transceiver module such as, but not limited to, a network interface card, a transceiver, etc., to implement communication between the computing device and other devices or a communication network.

[0066] Embodiment 4 This embodiment also provides a computing device cluster. The computing device cluster includes at least one computing device. The computing device can be a server, such as a central server, an edge server, or a local server in a local data center. In some embodiments, the computing device can also be a terminal device such as a desktop computer, a laptop computer, or a smart phone.

[0067] The computing device cluster includes at least one computing device. Instructions for performing the methods and functions of the computing device involved in any of the above embodiments can be stored in the memories of one or more computing devices in the computing device cluster.

[0068] In some possible implementation manners, partial instructions for performing the methods and functions of the computing device involved in any of the above embodiments can also be stored separately in the memories of one or more computing devices in the computing device cluster. In other words, a combination of one or more computing devices can jointly execute the instructions for performing the methods and functions of the computing device.

[0069] It should be noted that the memories in different computing devices in the computing device cluster can store different instructions, respectively for performing partial functions of the device.

[0070] In some possible implementation manners, one or more computing devices in the computing device cluster can be connected through a network. Among them, the network can be a wide area network or a local area network, etc. Two computing devices are connected through the network. Specifically, they are connected to the network through the communication interfaces in each computing device.

[0071] The embodiments of the present disclosure also provide a computer program product containing instructions, which, when running on a computer, causes the computer to execute the methods and functions of the computing device involved in any of the above embodiments.

[0072] Embodiment 5 This embodiment also provides a computer-readable storage medium, on which computer instructions are stored. When a processor runs the instructions, the processor is caused to execute the methods and functions of the computing device involved in any of the above embodiments.

[0073] In general, the various embodiments of the present disclosure may be implemented in hardware or specific circuits, software, logic, or any combination thereof. Some aspects may be implemented in hardware, while other aspects may be implemented in firmware or software, which may be executed by a controller, a microprocessor, or other computing devices. Although the various aspects of the embodiments of the present disclosure are shown and described as block diagrams, flowcharts, or using some other graphical representation, it should be understood that the blocks, devices, systems, techniques, or methods described herein may be implemented as, by way of non-limiting example, hardware, software, firmware, specific circuits or logic, general-purpose hardware or a controller or other computing devices, or some combination thereof.

[0074] Embodiment 6 This embodiment provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer-executable instructions, such as instructions included in program modules, which are executed in a device on a target real or virtual processor to perform the processes / methods as referred to the accompanying drawings above. Generally, program modules include routines, programs, libraries, objects, classes, components, data structures, etc. that perform specific tasks or implement specific abstract data types. In various embodiments, the functions of program modules may be combined or divided as needed among program modules. The machine-executable instructions for program modules may be executed within local or distributed devices. In a distributed device, program modules may be located in local and remote storage media.

[0075] The computer program code for implementing the methods of the present disclosure may be written in one or more programming languages. The computer program code may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the program code is executed by the computer or other programmable data processing device, the functions / operations specified in the flowchart and / or block diagram are implemented. The program code may be executed entirely on the computer, partially on the computer, as a stand-alone software package, partially on the computer and partially on a remote computer, or entirely on a remote computer or server.

[0076] In the context of the present disclosure, the computer program code or related data may be carried by any suitable carrier such that the device, apparatus, or processor can perform the various processes and operations described above. Examples of carriers include signals, computer-readable media, and the like. Examples of signals may include electrical, optical, radio, acoustic, or other forms of propagated signals, such as carrier waves, infrared signals, etc.

[0077] A computer-readable medium can be any tangible medium that contains or stores a program for or related to an instruction execution system, apparatus, or device, or a data storage device such as a data center that contains one or more available media. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. More specific examples of computer-readable storage media include electrical connections with one or more wires, portable computer disks, hard disks, random access memories (RAMs), read-only memories (ROMs), erasable programmable read-only memories (EPROMs or flash memories), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0078] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included within the protection scope of the present application.

Claims

1. A heating pipeline leakage identification method based on secondary decomposition and BiLSTM, characterized in that: The following steps are involved: Obtain the measured pressure value of the heating pipeline within a preset time period, and use the obtained measured pressure value as input data; Perform VMD-EMD secondary decomposition on the input data to obtain the residual components; The obtained residual component is used as the input of the constructed BiLSTM network model to obtain the predicted pressure value; The obtained predicted pressure value is compared with the preset fixed warning threshold, and whether the heating pipeline is leaking is determined based on the comparison result.

2. According to claim 1, a heating pipeline leakage identification method based on secondary decomposition and BiLSTM is characterized in that: Perform VMD-EMD secondary decomposition on the input data to obtain the secondary decomposition components. The specific method is: The input data is decomposed using VMD to obtain the modal component IMF including the residual term; The residual term is decomposed using EMD to obtain the residual components.

3. A heating pipeline leakage identification method based on secondary decomposition and BiLSTM according to claim 1, characterized in that: Construct a BiLSTM network model. The specific method is: Obtain historical pressure signals under normal working conditions, leakage working conditions, and valve regulating working conditions, and use the historical pressure signals as original input data; Perform VMD-EMD secondary decomposition on the original input data to obtain the residual components; The constructed BiLSTM network model is trained using the obtained residual component to obtain a trained BiLSTM network model.

4. A heating pipeline leakage identification method based on secondary decomposition and BiLSTM according to claim 3, characterized in that: The preset fixed warning threshold can be obtained by: The constructed BiLSTM network model is trained using the residual component to obtain prediction data; Compare the obtained predicted data with the actual data and calculate the prediction error value; A fixed warning threshold is set according to the obtained prediction error value.

5. The method for identifying leakage of a heating pipeline based on secondary decomposition and BiLSTM according to claim 1 is characterized in that: The predicted pressure value is compared with the preset fixed warning threshold, and whether the heating pipeline is leaking is determined based on the comparison result. The specific method is: The obtained predicted pressure value and the measured pressure value are used to calculate the prediction error; The obtained prediction error is compared with a preset fixed warning threshold, wherein when the prediction error is greater than the preset fixed warning threshold, the heating pipeline leaks; otherwise, the heating pipeline is normal.

6. A heating pipeline leakage identification system based on secondary decomposition and BiLSTM, characterized in that: include: A pressure signal acquisition unit is used to acquire the measured pressure value of the heating pipeline within a preset time period and use the obtained measured pressure value as input data; The data decomposition unit is used to perform VMD-EMD secondary decomposition on the input data to obtain residual components; A model prediction unit is used to use the obtained residual component as the input of the constructed BiLSTM network model to obtain a predicted pressure value; The judgment unit is used to compare the obtained predicted pressure value with a preset fixed warning threshold value, and judge whether the heating pipeline is leaking according to the comparison result.

7. An electronic device, characterized in that: The electronic device comprises a processor and a memory, wherein the memory stores computer instructions, and when the computer instructions are executed by the processor, the electronic device executes the method according to any one of claims 1 to 5.

8. A computing device cluster, characterized in that: comprising at least one computing device, each computing device comprising a processor and a memory; The processor of the at least one computing device is configured to execute instructions stored in the memory of the at least one computing device, so that the computing device cluster executes the method according to any one of claims 1 to 5.

9. A computer program product, characterized in that The computer program product contains computer executable instructions, which implement the method according to any one of claims 1 to 5 when executed.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images