Transmission tower foundation differential settlement monitoring method and system based on LSTM

Abstract

The invention discloses a transmission tower foundation differential settlement monitoring method and system based on LSTM, and belongs to the technical field of transmission tower monitoring. The system comprises a monitoring unit, a data processing unit, an operation unit, a judgment unit and an early warning unit. The method comprises the steps that monitoring data of differential settlement of the transmission tower and monitoring data of the surrounding environment of the transmission tower are collected through a monitoring unit; a data processing unit is used for preprocessing the monitoring data to form model data; an LSTM long short-term memory network algorithm is applied through an arithmetic unit, and a prediction model for differential settlement of the transmission tower foundation is constructed through model data training; the displacement data of the transmission tower in the future time are predicted according to the prediction model, the predicted displacement data are compared with standard displacement data through the judgment unit, and early warning is conducted through the early warning unit according to the displacement offset. According to the invention, manual monitoring cost is reduced by means of artificial intelligence, huge monitoring data can be processed, and monitoring timeliness and accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power transmission tower monitoring, and in particular to a method and system for monitoring uneven settlement of a power transmission tower foundation based on LSTM. Background Art

[0002] Monitoring the uneven settlement of transmission tower foundations is a key link in ensuring the safety, reliability and sustainable operation of the transmission system. Timely detection and resolution of uneven settlement problems can reduce risks, reduce maintenance costs and ensure the continuity of power transmission. In terms of structural safety, transmission towers are an important part of the power transmission system, erecting power lines and equipment. In terms of safety, uneven settlement may lead to instability in the transmission tower structure, misalignment of the towers of the power lines, resulting in line breakage or poor contact, increasing the risk of overturning or collapse, causing power outages and power failures, and posing a threat to power supply safety. In terms of maintenance costs, if uneven settlement is not detected and resolved in a timely manner, the cost of subsequent repairs and maintenance will be higher.

[0003] At present, there are still the following deficiencies in monitoring the uneven settlement of transmission tower foundations: 1) Complex influencing factors: Monitoring uneven settlement requires consideration of multiple factors, such as soil conditions, meteorological conditions, and structural parameters. Traditional methods have difficulty in effectively dealing with the interaction of these complex factors. 2) The current high cost of manual monitoring: Manual monitoring requires a lot of manpower and time resources, especially for large-scale transmission tower systems. This may increase the cost of monitoring. 3) Lack of automated prediction: Traditional monitoring methods usually involve manual analysis and decision-making, lacking automation and predictability. This may cause the problem to be ignored until a serious situation occurs. 4) Difficulty in processing big data: Monitoring the uneven settlement of transmission tower foundations involves a large amount of data, including displacement, inclination, temperature and humidity, pressure, vibration, and sound. Traditional methods have difficulty in effectively processing and analyzing big data, especially in real time. 5) Insufficient precision and accuracy: Traditional methods may be affected by subjectivity and human errors, which reduces the precision and accuracy of monitoring data.

[0004] To this end, we designed a LSTM-based transmission tower foundation uneven settlement monitoring method and system to solve the above problems. Summary of the invention

[0005] The purpose of the present invention is to solve the problems existing in the prior art that there are too many factors affecting the monitoring of uneven settlement of transmission tower foundations, which are difficult to take into account comprehensively, the monitoring data is huge, the data processing is difficult, and it is impossible to efficiently realize the automatic prediction of uneven settlement of transmission tower foundations. The means of manual monitoring are susceptible to subjectivity and human errors, lack of accuracy, and high cost. Therefore, a method and system for monitoring uneven settlement of transmission tower foundations based on LSTM are proposed. The LSTM algorithm is used to process complex multi-factor data of large-scale time series, reduce the cost of manual monitoring, and improve the intelligence level and accuracy of monitoring uneven settlement of transmission tower foundations.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A method for monitoring uneven settlement of transmission tower foundation based on LSTM, comprising the following steps:

[0008] Step 1, collecting monitoring data of uneven settlement of transmission towers and the surrounding environment of transmission towers through a monitoring unit;

[0009] Step 2, using a data processing unit to pre-process the monitoring data to form model data;

[0010] Step 3, using the LSTM long short-term memory network algorithm through the computing unit, and using the model data training to build a prediction model for uneven settlement of the transmission tower foundation;

[0011] Step 4: predict the displacement data of the transmission tower in the future according to the prediction model, compare the predicted displacement data with the standard displacement data through the judgment unit, and issue a warning according to the displacement offset through the warning unit.

[0012] Further preferably, the monitoring unit includes a displacement sensor, an inclination sensor, a pressure sensor, a vibration sensor, a temperature sensor, a humidity sensor and a sound sensor; the displacement sensor collects the displacement change of the transmission tower, the pressure sensor collects the pressure change of the soil at the foundation of the transmission tower, the vibration sensor collects the amplitude and frequency of the vibration of the transmission tower body, the temperature sensor and the humidity sensor collect the temperature and humidity of the environment surrounding the transmission tower, and the sound sensor collects the sound emitted by the soil at the foundation of the transmission tower and the transmission tower structure.

[0013] Further preferably, in step 1, the monitoring data includes displacement data, tilt data, pressure data, vibration data, temperature data, humidity data and sound data, and the monitoring data is expressed by the following formula:

[0014] X(t)=[ΔS(t),θt),P(t),D(t),T(t),W(t),F(t)] (1)

[0015] Where X(t) is the monitoring data of the transmission tower collected at time t; ΔS(t) is the displacement data of the transmission tower collected at time t; θ(t) is the inclination data of the transmission tower body collected at time t, P(t) is the pressure data of the transmission tower foundation soil collected at time t, D(t) is the vibration data of the transmission tower body structure collected at time t, T(t) is the temperature data of the surrounding environment of the transmission tower collected at time t, W(t) is the humidity data of the surrounding environment of the transmission tower collected at time t, and F(t) is the sound data emitted by the transmission tower foundation soil and the transmission tower structure collected at time t.

[0016] Further preferably, in step 2, the monitoring data is preprocessed by a data processing unit:

[0017] For the displacement data ΔS(t) of the transmission tower, according to formula (1), the displacement data ΔS(t) is transformed by the data processing unit to obtain the model data, as shown below:

[0018] X′(t)=[ΔS′(t), θ′(t), P′(t), D′(t), T′(t), W′(t), F′(t)] (2)

[0019]

[0020] It is obtained that X′(t) is the model data collected for transmission tower monitoring after preprocessing at time t; ΔS′(t) is the model data of transmission tower displacement after preprocessing at time t, and n is the number of samples of ΔS(t); θ′(t) is the model data of transmission tower inclination after preprocessing at time t, P′(t) is the model data of soil pressure of transmission tower foundation after preprocessing at time t, D′(t) is the model data of transmission tower structure vibration after preprocessing at time t, T′(t) is the model data of ambient temperature around transmission tower after preprocessing at time t, W′(t) is the model data of ambient humidity around transmission tower after preprocessing at time t, and F′(t) is the model data of sound emitted by transmission tower foundation soil and transmission tower structure after preprocessing at time t.

[0021] Further preferably, in step 3, the LSTM long short-term memory network algorithm is used by the computing unit to train and construct a prediction model for uneven settlement of the transmission tower foundation using model data, including the following contents:

[0022] The model data formed after the data processing unit pre-processes the monitoring data is used as an input sample;

[0023] Input standard displacement data into the hidden layer;

[0024] Based on the input sample data and cell state information of the model data, the LSTM deep learning network model is trained. The training process is shown in formula (4):

[0025]

[0026] In the above formula, σ is the activation function of the prediction model of uneven settlement of transmission tower foundation; f t is the output of the forget gate of the input prediction model, W f 、b f is the forget gate matrix of the prediction model; i t is the output of the prediction model input gate, W i 、b i Input gate weight matrix for prediction model; C t-1 To predict the old cell state information of the model, The candidate state information added for the prediction model selection, C t is the cell information updated by the prediction model, W c 、b c is the neuron matrix corresponding to the model; O t is the output of the prediction model output gate, W o 、b o is the output gate matrix corresponding to the prediction model; h t is the output result of the prediction model;

[0027] According to the prediction model of uneven settlement of transmission tower foundation, the displacement data of transmission tower in the future is predicted, as shown in formula (5):

[0028] F C (t+Δt)=f(ΔS′(t), θ′(t), P′(t), D′(t), T′(t), W′(t), F′(t)) (5)

[0029] Where: f() is the prediction model of uneven settlement of transmission tower foundation. According to the prediction model, when ΔS′(t), θ′(t), P′(t), D′(t), T′(t), W′(t), F′(t) are input, the displacement data ΔS of the transmission tower at the future time t+Δt is predicted. y (t+Δt).

[0030] Further preferably, in step 4, the transmission tower displacement data at a future time is predicted according to the prediction model, the predicted displacement data is compared with the standard displacement data by the judgment unit, and an early warning is issued according to the displacement offset by the early warning unit, including the following steps:

[0031] Step 4.1: Calculate the displacement data of the standard transmission tower according to formula (6):

[0032]

[0033] In the formula, ΔS B is the displacement data of uneven settlement of transmission towers calculated at a fixed time each year, P N is the vertical load applied to the transmission tower foundation, L is the length of the transmission tower foundation; A is the area of ​​the foundation cross section of the transmission tower; E is the elastic modulus of the foundation soil;

[0034] Step 4.2: The displacement data of the uneven settlement of the transmission tower calculated at a fixed time each year is compared with the displacement data ΔS predicted at the future time t+Δt by the judgment unit. y (t+Δt) is compared and the difference is obtained. According to formula (7), the deviation of the uneven foundation settlement of the transmission tower at the future time t+Δt is calculated.

[0035] G(t+Δt)=ΔS B -ΔS y (t+Δt) (7)

[0036] Where G(t+Δt) is the displacement deviation value of the transmission tower;

[0037] Step 4.3: Give an early warning according to formula (8).

[0038]

[0039] Where, BJ is the alarm value of uneven settlement of transmission tower foundation;

[0040] The early warning unit will issue an early warning based on the displacement deviation value to monitor the uneven settlement of the transmission tower foundation. When the deviation is within 2 mm, the alarm value BJ of the uneven settlement of the transmission tower foundation is 1, the indicator light is green, and no measures are required; when the deviation is between 2 mm and 5 mm, the alarm value BJ of the uneven settlement of the transmission tower foundation is 2, the indicator light is yellow, and further monitoring and evaluation are required; when the deviation exceeds 5 mm, the alarm value BJ of the uneven settlement of the transmission tower foundation is 3, the indicator light is red, and remedial measures need to be taken immediately.

[0041] A system for monitoring uneven settlement of a transmission tower foundation based on LSTM, the system comprising:

[0042] Monitoring unit, used to collect monitoring data of uneven settlement of transmission towers and the surrounding environment of transmission towers;

[0043] A data processing unit pre-processes the monitoring data to form model data for training the model;

[0044] The computing unit uses the LSTM long short-term memory network algorithm and the model data training to build a prediction model for uneven settlement of transmission tower foundations;

[0045] A judgment unit, used to compare the predicted displacement data with the standard displacement data to obtain the displacement deviation value of the transmission tower;

[0046] The early warning unit sends early warning prompts to the system based on the monitoring situation of uneven settlement of the transmission tower foundation according to the displacement deviation value.

[0047] Further preferably, the monitoring unit includes a displacement sensor, an inclination sensor, a pressure sensor, a vibration sensor, a temperature sensor, a humidity sensor and a sound sensor; the displacement sensor and the pressure sensor are arranged at the base of the transmission tower to collect the displacement changes of the base of the transmission tower and the pressure changes of the soil; the vibration sensor is installed on the tower body structure of the transmission tower to collect the amplitude and frequency of the vibration of the transmission tower; the temperature sensor and the humidity sensor are installed on the transmission tower to collect the temperature and humidity of the surrounding environment of the transmission tower; the sound sensor is arranged on the soil of the foundation of the transmission tower and the structure of the transmission tower to collect the sound information emitted by the soil of the foundation of the transmission tower and the structure of the transmission tower.

[0048] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention collects monitoring data of factors affecting uneven settlement of transmission towers in multiple dimensions, and normalizes the monitoring data. Based on the artificial intelligence LSTM algorithm, it processes complex multi-factor data of large-scale time series, trains a prediction model for predicting the trend of uneven settlement of transmission tower foundations in the future, uses the trained prediction model to predict the development trend of uneven settlement of transmission towers, and compares the prediction results with conventional standards to formulate an early warning system for uneven settlement of transmission tower foundations. The present invention reduces the cost of manual monitoring by means of artificial intelligence, effectively avoids the misjudgment of manual monitoring by personal subjective factors, improves the timeliness and accuracy of monitoring, comprehensively considers the factors affecting the monitoring of uneven settlement of transmission tower foundations, and comprehensively and efficiently realizes the monitoring of uneven settlement trends of transmission tower foundations. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A schematic diagram of the execution flow of a method for monitoring uneven settlement of a transmission tower foundation based on LSTM proposed by the present invention;

[0050] Figure 2 This is a schematic diagram of the logic state of the LSTM long short-term memory network algorithm used in an embodiment of the present invention;

[0051] Figure 3The figure is a schematic diagram of the system composition of the LSTM-based transmission tower foundation uneven settlement monitoring method applied in an embodiment of the present invention. DETAILED DESCRIPTION

[0052] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0053] A method for monitoring uneven settlement of transmission tower foundation based on LSTM. The process of executing the method is shown as follows: Figure 1 As shown, the following steps are included:

[0054] Step 1, propose a monitoring data collection technology for factors related to uneven settlement of transmission tower foundations, and collect monitoring data of uneven settlement of transmission towers and the surrounding environment of transmission towers through a monitoring unit;

[0055] The monitoring units used in the data collection in this embodiment include displacement sensors, inclination sensors, pressure sensors, vibration sensors, temperature sensors, humidity sensors and sound sensors; the displacement sensor collects the displacement changes of the transmission tower, and a rangefinder or a total station can also be used; the pressure sensor collects the pressure changes of the soil at the foundation of the transmission tower, the vibration sensor collects the amplitude and frequency of the vibration of the transmission tower body, the temperature sensor and the humidity sensor collect the temperature and humidity of the environment around the transmission tower, and the sound sensor collects the sound emitted by the soil at the foundation of the transmission tower and the transmission tower structure.

[0056] The monitoring data collected by the monitoring unit include displacement data, tilt data, pressure data, vibration data, temperature data, humidity data and sound data. The monitoring data is expressed in the following formula:

[0057] X(t)=[ΔS(t),θt),P(t),D(t),T(t),W(t),F(t)] (1)

[0058] Where X(t) is the monitoring data of the transmission tower collected at time t; ΔS(t) is the displacement data of the transmission tower collected at time t; θ(t) is the inclination data of the transmission tower body collected at time t, P(t) is the pressure data of the transmission tower foundation soil collected at time t, D(t) is the vibration data of the transmission tower body structure collected at time t, T(t) is the temperature data of the surrounding environment of the transmission tower collected at time t, W(t) is the humidity data of the surrounding environment of the transmission tower collected at time t, and F(t) is the sound information data emitted by the transmission tower foundation soil and the transmission tower structure collected at time t.

[0059] Step 2 mainly uses the data processing unit to pre-process the monitoring data to form model data, providing data support for the uneven settlement monitoring of transmission tower foundations. Since the collected uneven settlement monitoring data of transmission tower foundations have different units and dimensions, it is necessary to pre-process the data to normalize it before using artificial intelligence to analyze it, so as to facilitate model training and subsequent online monitoring.

[0060] Specifically, preprocessing the monitoring data using the data processing unit includes the following:

[0061] For the displacement data ΔS(t) of the transmission tower, according to formula (1), the displacement data ΔS(t) is transformed by the data processing unit to obtain the model data, as shown below:

[0062] X′(t)=[ΔS′(t), θ′(t), P′(t), D′(t), T′(t), W′(t), F′(t)] (2)

[0063]

[0064] It is obtained that X'(t) is the model data collected for transmission tower monitoring after preprocessing at time t; ΔS'(t) is the model data of the transmission tower displacement after preprocessing at time t, and n is the number of samples of ΔS(t); θ'(t) is the model data of the transmission tower body inclination after preprocessing at time t, P'(t) is the model data of the soil pressure of the transmission tower foundation after preprocessing at time t, D'(t) is the model data of the transmission tower body structure vibration after preprocessing at time t, T'(t) is the model data of the ambient temperature around the transmission tower after preprocessing at time t, W'(t) is the model data of the ambient humidity around the transmission tower after preprocessing at time t, and F'(t) is the model data of the sound emitted by the transmission tower foundation soil and the transmission tower structure after preprocessing at time t.

[0065] Step 3: Use the LSTM (Long Short-Term Memory, LSTM) long short-term memory network algorithm through the computing unit, use the model data training to build a prediction model for uneven settlement of the transmission tower foundation, and predict the trend of uneven settlement of the transmission tower foundation in the future.

[0066] refer to Figure 2 , through the operation unit using LSTM long short-term memory network algorithm, using model data training to build a prediction model for uneven settlement of transmission tower foundation, including the following contents:

[0067] Step 3.1, the model data formed after the data processing unit preprocesses the monitoring data is used as the input sample, and the input sample is X(t) = [ΔS(t), θ(t), P′(t), D′(t), T′(t), W′(t), F′(t)].

[0068] Step 3.2: Input the standard displacement data (h t-1 =F C (t-1)), the standard displacement data is based on the displacement data of uneven settlement of transmission towers calculated at a fixed time every year.

[0069] Step 3.3: Based on the input sample data and cell state information of the model data, the LSTM deep learning network model is trained. The training process is shown in formula (4):

[0070]

[0071] In the above formula, σ is the activation function of the prediction model of uneven settlement of transmission tower foundation; f t is the output of the forget gate of the input prediction model, W f 、b f is the forget gate matrix of the prediction model; i t is the output of the prediction model input gate, W i 、b i Input gate weight matrix for prediction model; C t-1 To predict the old cell state information of the model, The candidate state information added for the prediction model selection, C t is the cell information updated by the prediction model, W c 、b c is the neuron matrix corresponding to the model; O t is the output of the prediction model output gate, W o 、b o is the output gate matrix corresponding to the prediction model; h t is the output result of the prediction model;

[0072] Step 3.4, based on the prediction model of uneven settlement of the transmission tower foundation, predict the displacement data of the transmission tower in the future, as shown in formula (5):

[0073] F C (t+Δt)=f(ΔS′(t), θ′(t), P′(t), D′(t), T′(t), W′(t), F′(t)) (5)

[0074] Where: f() is the prediction model of uneven settlement of transmission tower foundation. According to the prediction model, when ΔS′(t), θ′(t), P′(t), D′(t), T′(t), W′(t), F′(t) are input, the displacement data ΔS of the transmission tower at the future time t+Δt is predicted. y (t+Δt), that is, the output of the prediction model at the future time t+Δt is h t+Δt =F C (t+Δt)=[ΔS y (t+Δt)].

[0075] Step 4: predict the displacement data of the transmission tower in the future according to the prediction model, and compare the predicted displacement data with the standard displacement data (the standard displacement data refers to the displacement data of the uneven settlement of the transmission tower calculated at a fixed time every year, and the displacement data is known according to the measurement) through the judgment unit to obtain the displacement deviation value, thereby obtaining the displacement offset, and then issue an early warning according to the displacement offset through the early warning unit. The specific operation includes the following steps:

[0076] Step 4.1, calculate the displacement data of the standard transmission tower according to formula (6),

[0077]

[0078] In the formula, ΔS B is the displacement data of uneven settlement of transmission towers calculated at a fixed time each year, P N is the vertical load applied to the transmission tower foundation, L is the length of the transmission tower foundation; A is the area of ​​the foundation cross section of the transmission tower; E is the elastic modulus of the foundation soil;

[0079] Step 4.2: The displacement data of the uneven settlement of the transmission tower calculated at a fixed time each year is compared with the displacement data ΔS predicted at the future time t+Δt by the judgment unit. y (t+Δt) is compared and the difference is obtained. According to formula (7), the deviation of the uneven foundation settlement of the transmission tower at the future time t+Δt is calculated.

[0080] G(t+Δt)=ΔS B -ΔS y (t+Δt) (7)

[0081] Where G(t+Δt) is the displacement deviation value of the transmission tower;

[0082] Step 4.3: Give an early warning according to formula (8).

[0083]

[0084] Where, BJ is the alarm value of uneven settlement of transmission tower foundation;

[0085] The early warning unit issues early warning based on the displacement deviation value for monitoring uneven settlement of the transmission tower foundation:

[0086] 1) When the deviation is within 2 mm, the alarm value BJ of uneven settlement of the transmission tower foundation is 1, the indicator light is green, and no measures are required.

[0087] 2) When the deviation is between 2 mm and 5 mm, the alarm value BJ of uneven settlement of the transmission tower foundation is 2, and the indicator light is yellow, which requires further monitoring and evaluation.

[0088] 3) When the deviation exceeds 5 mm, the alarm value BJ of uneven settlement of the transmission tower foundation is 3, the indicator light is red, and remedial measures need to be taken immediately.

[0089] like Figure 3 As shown, the present invention proposes a system for monitoring uneven settlement of transmission tower foundation based on LSTM based on the LSTM-based transmission tower foundation, which includes: a monitoring unit, a data processing unit, a calculation unit, a judgment unit and an early warning unit.

[0090] The monitoring unit is used to collect monitoring data on uneven settlement of transmission towers and the surrounding environment of transmission towers; the data processing unit pre-processes the monitoring data to form model data for training the model; the operation unit uses the LSTM long short-term memory network algorithm to train and construct a prediction model for uneven settlement of transmission tower foundations using model data; the judgment unit is used to compare and differ the predicted displacement data with the standard displacement data to obtain the displacement deviation value of the transmission tower; the early warning unit sends an early warning prompt to the system based on the monitoring situation of uneven settlement of the transmission tower foundation according to the displacement deviation value.

[0091] Specifically, in this embodiment, the monitoring unit includes a displacement sensor, a tilt sensor, a pressure sensor, a vibration sensor, a temperature sensor, a humidity sensor, and a sound sensor.

[0092] Displacement sensors and pressure sensors are installed at the base of transmission towers to collect displacement changes of the base of transmission towers and pressure changes of the soil.

[0093] The vibration sensor is installed on the tower structure of the transmission tower to collect the amplitude and frequency of the transmission tower vibration.

[0094] Temperature sensors and humidity sensors are installed on transmission towers to collect the temperature and humidity of the surrounding environment of the transmission towers.

[0095] The sound sensor is arranged on the foundation soil of the electric pole tower and the transmission pole tower structure to collect the sound information emitted by the foundation soil of the transmission pole tower and the transmission pole tower structure.

[0096] This embodiment also provides a storage medium containing computer executable instructions, which are used to execute the LSTM-based transmission tower foundation uneven settlement monitoring method when the system is running.

[0097] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, a device or a device or used in combination with it.

[0098] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0099] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination of the foregoing.

[0100] The computer program code for executing the LSTM-based transmission tower foundation uneven settlement monitoring method of the present invention can be written in one or more programming languages ​​or a combination thereof, and the programming language includes an object-oriented programming language, such as Java, Smalltalk, C++, and also includes a conventional procedural programming language, such as "C" language or a similar programming language. The program code can be executed entirely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0101] Of course, the storage medium containing computer executable instructions provided in this embodiment is not limited to the above method operations, and the computer executable instructions can also execute any related operations in the LSTM-based transmission tower foundation uneven settlement monitoring method of the present invention.

[0102] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for monitoring uneven settlement of transmission tower foundation based on LSTM, characterized in that: The following steps are involved: Step 1, collecting monitoring data of uneven settlement of transmission towers and the surrounding environment of transmission towers through a monitoring unit; Step 2, using a data processing unit to pre-process the monitoring data to form model data; Step 3, using the LSTM long short-term memory network algorithm through the computing unit, and using the model data training to build a prediction model for uneven settlement of the transmission tower foundation; Step 4: predict the displacement data of the transmission tower in the future according to the prediction model, compare the predicted displacement data with the standard displacement data through the judgment unit, and issue a warning according to the displacement offset through the warning unit.

2. According to the LSTM-based transmission tower foundation uneven settlement monitoring method of claim 1, it is characterized in that: The monitoring unit includes a displacement sensor, an inclination sensor, a pressure sensor, a vibration sensor, a temperature sensor, a humidity sensor and a sound sensor; the displacement sensor collects the displacement change of the transmission tower, the pressure sensor collects the pressure change of the transmission tower foundation soil, the vibration sensor collects the amplitude and frequency of the transmission tower body vibration, the temperature sensor and humidity sensor collect the temperature and humidity of the environment around the transmission tower, and the sound sensor collects the sound emitted by the transmission tower foundation soil and the transmission tower structure.

3. The method for monitoring uneven settlement of transmission tower foundation based on LSTM according to claim 1 is characterized in that: In step 1, the monitoring data includes displacement data, tilt data, pressure data, vibration data, temperature data, humidity data and sound data, and the monitoring data is expressed by the following formula: X(t)=[ΔS(t), θ(t), P(t), D(t), T(t), W(t), F(t)] (1) Where X(t) is the monitoring data of the transmission tower collected at time t; ΔS(t) is the displacement data of the transmission tower collected at time t; θ(t) is the inclination data of the transmission tower body collected at time t, P(t) is the pressure data of the transmission tower foundation soil collected at time t, D(t) is the vibration data of the transmission tower body structure collected at time t, T(t) is the temperature data of the surrounding environment of the transmission tower collected at time t, W(t) is the humidity data of the surrounding environment of the transmission tower collected at time t, and F(t) is the sound data emitted by the transmission tower foundation soil and the transmission tower structure collected at time t.

4. The method for monitoring uneven settlement of transmission tower foundation based on LSTM according to claim 1, characterized in that: In step 2, the monitoring data is preprocessed by using a data processing unit: For the displacement data ΔS(t) of the transmission tower, according to formula (1), the displacement data ΔS(t) is transformed by the data processing unit to obtain the model data, as shown below: X′(t)=[ΔS′(t), θ′(t), P′(t), D′(t), T′(t), W′(t), F′(t)] (2) It is obtained that X'(t) is the model data collected for transmission tower monitoring after preprocessing at time t; ΔS'(t) is the model data of the transmission tower displacement after preprocessing at time t, and n is the number of samples of ΔS(t); θ'(t) is the model data of the transmission tower body inclination after preprocessing at time t, P'(t) is the model data of the soil pressure of the transmission tower foundation after preprocessing at time t, D'(t) is the model data of the transmission tower body structure vibration after preprocessing at time t, T'(t) is the model data of the ambient temperature of the transmission tower after preprocessing at time t, W'(t) is the model data of the ambient humidity of the transmission tower after preprocessing at time t, and F'(t) is the model data of the sound emitted by the transmission tower foundation soil and the transmission tower structure after preprocessing at time t.

5. The method for monitoring uneven settlement of transmission tower foundation based on LSTM according to claim 1, characterized in that: In step 3, the LSTM long short-term memory network algorithm is used by the computing unit to train the model data to build a prediction model for uneven settlement of the transmission tower foundation, including the following contents: The model data formed after the data processing unit preprocesses the monitoring data is used as the input sample; the standard displacement data is input into the hidden layer, and the LSTM deep learning network model is trained based on the input sample data and cell state information of the model data. The training process is shown in formula (4): In the above formula, σ is the activation function of the prediction model of uneven settlement of transmission tower foundation; f t is the output of the forget gate of the input prediction model, W f 、b f is the forget gate matrix of the prediction model; i t is the output of the prediction model input gate, W i 、b i Input gate weight matrix for prediction model; C t-1 To predict the old cell state information of the model, The candidate state information added for the prediction model selection, C t is the cell information updated by the prediction model, W c 、b C is the neuron matrix corresponding to the model; o t is the output of the prediction model output gate, W o 、b o is the output gate matrix corresponding to the prediction model; h t is the output result of the prediction model; According to the prediction model of uneven settlement of transmission tower foundation, the displacement data of transmission tower in the future is predicted, as shown in formula (5): F C (t+Δt)=f(ΔS′(t),θ′(t),P′(t),D′(t),T′(t),W′(t),F′(t)) (5) Where: f() is the prediction model of uneven settlement of transmission tower foundation. According to the prediction model, when ΔS'(t), θ'(t), P'(t), D'(t), T'(t), W'(t), F'(t) are input, the displacement data ΔS of the transmission tower at the future time t+Δt is predicted. y (t+Δt).

6. The LSTM-based method for monitoring uneven settlement of transmission tower foundation according to claim 1, characterized in that: In step 4, the transmission tower displacement data of the future time is predicted according to the prediction model, the predicted displacement data is compared with the standard displacement data through the judgment unit, and the early warning unit issues an early warning according to the displacement offset, including the following steps: Step 4.1: Calculate the displacement data of the standard transmission tower according to formula (6): In the formula, ΔS B is the displacement data of uneven settlement of transmission towers calculated at a fixed time each year, P N is the vertical load applied to the transmission tower foundation, L is the length of the transmission tower foundation; A is the area of ​​the foundation cross section of the transmission tower; E is the elastic modulus of the foundation soil; Step 4.2: The displacement data of the uneven settlement of the transmission tower calculated at a fixed time each year is compared with the displacement data ΔS predicted at the future time t+Δt by the judgment unit. y (t+Δt) is compared and the difference is obtained. According to formula (7), the deviation of the uneven foundation settlement of the transmission tower at the future time t+Δt is calculated. G(t+Δt)=ΔS B -ΔS y (t+Δt) (7) Where G(t+Δt) is the displacement deviation value of the transmission tower; Step 4.3: Give an early warning according to formula (8). Where, BJ is the alarm value of uneven settlement of transmission tower foundation; The early warning unit will issue an early warning based on the displacement deviation value to monitor the uneven settlement of the transmission tower foundation. When the deviation is within 2 mm, the alarm value BJ of the uneven settlement of the transmission tower foundation is 1, the indicator light is green, and no measures are required; when the deviation is between 2 mm and 5 mm, the alarm value BJ of the uneven settlement of the transmission tower foundation is 2, the indicator light is yellow, and further monitoring and evaluation are required; when the deviation exceeds 5 mm, the alarm value BJ of the uneven settlement of the transmission tower foundation is 3, the indicator light is red, and remedial measures need to be taken immediately.

7. A system applied to the LSTM-based transmission tower foundation uneven settlement monitoring method according to claims 1-6, characterized in that: The system comprises: Monitoring unit, used to collect monitoring data of uneven settlement of transmission towers and the surrounding environment of transmission towers; A data processing unit pre-processes the monitoring data to form model data for training the model; The computing unit uses the LSTM long short-term memory network algorithm and the model data training to build a prediction model for uneven settlement of transmission tower foundations; A judgment unit, used to compare the predicted displacement data with the standard displacement data to obtain the displacement deviation value of the transmission tower; The early warning unit sends early warning prompts to the system based on the monitoring situation of uneven settlement of the transmission tower foundation according to the displacement deviation value.

8. The system according to claim 7, characterized in that The monitoring unit includes a displacement sensor, an inclination sensor, a pressure sensor, a vibration sensor, a temperature sensor, a humidity sensor and a sound sensor; the displacement sensor and the pressure sensor are arranged at the base of the transmission tower to collect the displacement change of the base of the transmission tower and the pressure change of the soil; the vibration sensor is installed on the tower body structure of the transmission tower to collect the amplitude and frequency of the vibration of the transmission tower; the temperature sensor and the humidity sensor are installed on the transmission tower to collect the temperature and humidity of the environment around the transmission tower; the sound sensor is arranged on the soil of the base of the transmission tower and the structure of the transmission tower to collect the sound information emitted by the soil of the base of the transmission tower and the structure of the transmission tower.

Images