WO2022198899A1 - 一种海底电缆故障诊断方法、装置及设备 - Google Patents

一种海底电缆故障诊断方法、装置及设备 Download PDF

Info

Publication number
WO2022198899A1
WO2022198899A1 PCT/CN2021/113163 CN2021113163W WO2022198899A1 WO 2022198899 A1 WO2022198899 A1 WO 2022198899A1 CN 2021113163 W CN2021113163 W CN 2021113163W WO 2022198899 A1 WO2022198899 A1 WO 2022198899A1
Authority
WO
WIPO (PCT)
Prior art keywords
submarine cable
fault diagnosis
target
data
fault
Prior art date
Application number
PCT/CN2021/113163
Other languages
English (en)
French (fr)
Inventor
王磊
杨鹏
王学彬
赵春雷
李春晓
宋文乐
Original Assignee
国网河北省电力有限公司沧州供电分公司
国网河北省电力有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 国网河北省电力有限公司沧州供电分公司, 国网河北省电力有限公司 filed Critical 国网河北省电力有限公司沧州供电分公司
Priority to US17/599,552 priority Critical patent/US11774485B2/en
Publication of WO2022198899A1 publication Critical patent/WO2022198899A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/08—Locating faults in cables, transmission lines, or networks
    • G01R31/081—Locating faults in cables, transmission lines, or networks according to type of conductors
    • G01R31/083—Locating faults in cables, transmission lines, or networks according to type of conductors in cables, e.g. underground
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/08—Locating faults in cables, transmission lines, or networks
    • G01R31/088—Aspects of digital computing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/047—Probabilistic or stochastic networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/086—Learning methods using evolutionary algorithms, e.g. genetic algorithms or genetic programming
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/12—Computing arrangements based on biological models using genetic models
    • G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming

Definitions

  • the invention belongs to the technical field of submarine cable maintenance, and in particular relates to a method, device and equipment for fault diagnosis of a submarine cable.
  • the present invention provides a submarine cable fault diagnosis method, device and equipment to solve the problem of monitoring the state of the submarine cable and determining the fault type in time.
  • a first aspect of the embodiments of the present invention provides a method for diagnosing a submarine cable fault, including:
  • the deployment data and sensing data are input into the trained fault diagnosis model, and the diagnosis result of the target submarine cable is obtained, wherein the fault diagnosis model is a probabilistic neural network model, and the diagnosis result includes the fault type.
  • the deployment data and sensing data into the trained fault diagnosis model before inputting the deployment data and sensing data into the trained fault diagnosis model, further include:
  • the initial fault diagnosis model is trained based on the differential evolution algorithm and training samples to obtain the trained fault diagnosis model.
  • the initial fault diagnosis model is trained based on the differential evolution algorithm and training samples to obtain the trained fault diagnosis model including:
  • the smoothing parameters in the initial fault diagnosis model are optimized by the differential evolution algorithm to obtain the target smoothing parameters.
  • the item to be optimized in the differential evolution algorithm is the smoothing parameter
  • the objective function is the fault diagnosis model obtained by fault diagnosis of the training samples.
  • the fault diagnosis model using target smoothing parameters is trained based on the training samples to obtain a trained fault diagnosis model.
  • the smoothing parameters in the initial fault diagnosis model are optimized by the differential evolution algorithm to obtain the target smoothing parameters including:
  • the optimal individual in the offspring population is determined as the target smoothing parameter; otherwise, the offspring population is taken as the new parent population, and jumps to "In the parent population" Randomly select a specified number of individuals as mutation objects, and perform evolutionary operations on the parent population based on the mutation objects and the objective function".
  • the method further includes:
  • twin database includes deployment data, sensing data and fault types of historically faulty submarine cables, and the training samples are based on the deployment data, sensing data and fault types of historically faulty submarine cables generate.
  • the method also includes:
  • the digital twin is updated.
  • acquiring the deployment data and sensing data of the target submarine cable includes:
  • the sensing equipment includes at least one of the following: ship monitoring equipment, optical fiber disturbance monitoring equipment, temperature detection equipment and strain monitoring equipment;
  • the deployment data includes environmental information and/or cable orientation information of the target submarine cable, and the sensing data includes at least one of the following:
  • the method further includes:
  • an alarm message is sent to the target terminal device, so that the target terminal device displays the alarm message, wherein the alarm message includes at least one of the following: the identification of the target submarine cable, Fault type and location message.
  • a second aspect of the embodiments of the present invention provides a submarine cable fault diagnosis device, including:
  • an acquisition module for acquiring deployment data and sensing data of the target submarine cable
  • the diagnosis module is used for inputting deployment data and sensing data into the trained fault diagnosis model to obtain the diagnosis result of the target submarine cable, wherein the fault diagnosis model is a probabilistic neural network model, and the diagnosis result includes the fault type.
  • the device further includes:
  • a model building module for building an initial fault diagnosis model before inputting deployment data and sensing data into the trained fault diagnosis model
  • the model training module is used to train the initial fault diagnosis model based on the differential evolution algorithm and training samples to obtain the trained fault diagnosis model.
  • model training module includes:
  • the smoothing parameter determination unit is used to optimize the smoothing parameters in the initial fault diagnosis model through the differential evolution algorithm to obtain the target smoothing parameter, wherein the item to be optimized in the differential evolution algorithm is the smoothing parameter, and the objective function is the fault diagnosis model pair.
  • the sample training unit is used to train the fault diagnosis model using the target smoothing parameter based on the training samples, so as to obtain the trained fault diagnosis model.
  • the smoothing parameter determination unit is also used to:
  • the optimal individual in the offspring population is determined as the target smoothing parameter; otherwise, the offspring population is taken as the new parent population, and jumps to "In the parent population" Randomly select a specified number of individuals as mutation objects, and perform evolutionary operations on the parent population based on the mutation objects and the objective function".
  • model training module further includes:
  • the sample acquisition unit is used to obtain training samples from the twin database before training the initial fault diagnosis model based on the differential evolution algorithm and the training samples to obtain the trained fault diagnosis model, wherein the twin database includes historical faults
  • the deployment data, sensing data, and fault types of submarine cables, and the training samples are generated based on the deployment data, sensing data, and fault types of historically faulty submarine cables.
  • the get module is also used to:
  • the digital twin is updated.
  • the get module is also used to:
  • the sensing equipment includes at least one of the following: ship monitoring equipment, optical fiber disturbance monitoring equipment, temperature detection equipment and strain monitoring equipment;
  • the deployment data includes environmental information and/or cable orientation information of the target submarine cable, and the sensing data includes at least one of the following:
  • the diagnosis result is stored in the twin database
  • an alarm message is sent to the target terminal device, so that the target terminal device displays the alarm message, wherein the alarm message includes at least one of the following: the identification of the target submarine cable, Fault type and location message.
  • a third aspect of the embodiments of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, when the processor executes the computer program The steps of implementing the method for diagnosing a submarine cable fault according to any one of the above.
  • a fourth aspect of the embodiments of the present invention provides a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the submarine cable fault diagnosis according to any one of the above is implemented steps of the method.
  • the present invention has the following beneficial effects:
  • the invention provides a fault diagnosis method for a submarine cable, comprising: acquiring deployment data and sensing data of a target submarine cable; inputting the deployment data and the sensing data into a trained fault diagnosis model to obtain the target submarine cable A cable diagnosis result, wherein the fault diagnosis model is a probabilistic neural network model, and the diagnosis result includes a fault type.
  • the invention uses the trained probability neural network model as the fault diagnosis model to diagnose the fault of the target submarine cable, and can quickly and accurately obtain the fault type of the target submarine cable.
  • Fig. 1 is the realization flow chart of the submarine cable fault diagnosis method provided by the embodiment of the present invention.
  • FIG. 2 is a schematic structural diagram of a submarine cable fault diagnosis device provided by an embodiment of the present invention.
  • FIG. 3 is a schematic diagram of an electronic device provided by an embodiment of the present invention.
  • FIG. 1 shows a flowchart of the implementation of the method for diagnosing a submarine cable fault provided by an embodiment of the present invention, which is described in detail as follows:
  • Step 101 obtaining deployment data and sensing data of the target submarine cable
  • the target submarine cable can be determined according to the specified cable position and the number of cables.
  • Step 102 Input the deployment data and the sensing data into the trained fault diagnosis model to obtain a diagnosis result of the target submarine cable, wherein the fault diagnosis model is a probabilistic neural network model, and the diagnosis result includes the fault type.
  • the fault diagnosis model is a probabilistic neural network model, and the diagnosis result includes the fault type.
  • a probabilistic neural network model is used as the fault diagnosis model.
  • the probabilistic neural network is a feedforward neural network, which has the advantages of a simple structure and a fast network training process.
  • the probabilistic neural network in the submarine cable fault diagnosis model consists of four layers: input layer, model layer, summation layer and output layer, which can process data in real time.
  • the deployment data and sensing data into the trained fault diagnosis model before inputting the deployment data and sensing data into the trained fault diagnosis model, further include:
  • the probability density function estimation formula of the probability neural network application in the submarine cable fault diagnosis model is as follows:
  • X is the fault sample to be judged
  • X ai is the ith training vector of the fault mode
  • P is the dimension of the fault data sample
  • m is the number of training samples of the fault mode
  • ⁇ is the smoothing parameter, and its value is suitable for classification
  • Results play a key role and are usually given by humans empirically.
  • the initial fault diagnosis model is trained based on the differential evolution algorithm and training samples to obtain the trained fault diagnosis model.
  • training the initial fault diagnosis model based on the differential evolution algorithm and the training samples to obtain the trained fault diagnosis model includes the following steps:
  • Step 1 Optimize the smoothing parameters in the initial fault diagnosis model through the differential evolution algorithm to obtain the target smoothing parameters, wherein the item to be optimized in the differential evolution algorithm is the smoothing parameter, and the objective function is the fault diagnosis model to fault the training samples.
  • the smoothing parameters in the initial fault diagnosis model are optimized by the differential evolution algorithm, so as to improve the classification speed and accuracy of the fault diagnosis model.
  • the smoothing parameters in the initial fault diagnosis model are optimized to obtain the target smoothing parameters including:
  • the optimal individual in the offspring population is determined as the target smoothing parameter; otherwise, the offspring population is taken as the new parent population, and jumps to "In the parent population" Randomly select a specified number of individuals as mutation objects, and perform evolutionary operations on the parent population based on the mutation objects and the objective function".
  • the differential evolution algorithm is used to optimize the smoothing parameters.
  • the differential evolution algorithm is a random heuristic search algorithm.
  • the operating parameters used in this embodiment include scaling factor F, crossover factor CR, population size M and maximum evolutionary algebra G.
  • the evolutionary operation consists of three processes of mutation, crossover and selection, and specifically includes the following steps:
  • rand ij (0,1) is a random decimal generated between [0, 1]
  • x ij (0) is the initial dataset.
  • x P2j (t)-x P3j (t) is the differentiation vector
  • p 1 , p 2 , p 3 are the serial numbers of individuals in the population
  • x b1j (t) is the best individual in the population in the current generation
  • h ij (t+1) refers to generating an individual for the t+1th iteration based on the optimal individual obtained from the t-th iteration.
  • the purpose of the crossover operation is to increase the diversity of the population, and the operation is:
  • randl ij is a random decimal between [0, 1]
  • CR is the crossover probability
  • the value range of CR is [0, 1]
  • v ij (t+1) refers to the t+1th iteration process
  • An individual is generated in , and a new sample is generated by replacing the original x ij (t) through the crossover operation, where x ij (t) refers to the jth element of the ith category in the tth iteration process.
  • the selection operation is used to determine whether the target vector x i1 (t) can become the next generation of the population, compare the target vector x i1 (t) with the objective function value corresponding to the test vector v i1 (t+1), and keep the target Individuals with smaller function values:
  • f(v i1 (t+1)) represents the objective function value corresponding to v i1 (t+1)
  • f(x i1 (t)) represents the objective function value corresponding to x i1 (t)
  • vi ( t) +1) means to operate on v ij (t+1) in step (3), that is, to replace each element in v i to generate an individual.
  • the differential evolution algorithm used in this embodiment can quickly find the optimal solution of the smoothing parameters, and can also improve the accuracy of the optimal solution, thereby improving the speed and accuracy of fault diagnosis of submarine cables.
  • Step 2 Train the fault diagnosis model using the target smoothing parameter based on the training samples to obtain a trained fault diagnosis model.
  • this embodiment further includes:
  • the specified submarine cable may be a submarine cable in a specified area, which may be the same as or different from the target submarine cable for fault diagnosis;
  • the engineering database includes the data of the submarine cable during construction, such as the location of the submarine cable. Topography, water depth, topography, geological conditions, hydrological data of the environment, as well as the direction and crossing data of submarine cables.
  • the digital twin is updated.
  • the physical entity is composed of an actual submarine cable and a sensor system.
  • the sensor collects the operation data of the submarine cable in real time and transmits it to the twin database for storage.
  • the digital twin is the digital mapping of the physical entity in the virtual space. Real-time monitoring of the running status of the submarine cable entity, and the simulation data is uploaded to the database in real time; the connection between each part can realize the dynamic flow of data, real-time update and iteration of data, and optimize the simulated digital twin to make it closer to the seabed Cable entity.
  • the maintenance plan of the submarine cable can be optimized and the probability of failure can be reduced.
  • To establish a digital twin of a specified submarine cable is to establish a high-fidelity model of the specified submarine cable in a virtual space, and to specify a digital description of the physical entity of the submarine cable in the virtual space, which uses digital models and reliable perception to transmit data to information in real time. layer to realize intelligent perception and real-time interaction.
  • the digital twin in this embodiment integrates and fuses four-layer models of geometry, physics, behavior, and rules, and can collect fault data in real time, simulate the actual operation state of a specified submarine cable, and perform multi-dimensional and multi-time scale analysis on physical entities. High-fidelity descriptions and real-time mapping can be used to simulate, monitor, and diagnose the state and behavior of physical entities in real-world environments. Based on the sensing data collected by the sensing device in real time, the digital twin can be updated to make the digital twin closer to the specified submarine cable entity.
  • the construction of the specified submarine cable digital twin provides a data-driven way to comprehensively describe the evolution process of complex uncertainties such as the aging process of equipment in different environments, the role of human operation and maintenance behavior, and interaction with the environment. It helps to realize the full life cycle management of submarine cables, effectively reduce the operating costs of submarine cables, and improve the service life of submarine cables.
  • the internal state information of complex links that are difficult to obtain by conventional means can be visually presented to the operation and maintenance personnel through technical means such as virtual reality and augmented reality. Detection and diagnosis provide important means, and field personnel can quickly locate fault points and implement repairs based on the information provided by the digital twin;
  • the method further includes:
  • twin database includes deployment data, sensing data and fault types of historically faulty submarine cables, and the training samples are based on the deployment data, sensing data and fault types of historically faulty submarine cables generate.
  • the manufacturing, principle, and components of the submarine cable entity, as well as all data information, real-time data, and simulation data of the virtual submarine cable that constitute the geometric, physical, behavioral and rule models of the digital twin are stored in the twin database.
  • the training samples are obtained from the database to train the fault diagnosis model, so that the fault diagnosis model can obtain more accurate diagnosis results.
  • acquiring the deployment data and sensing data of the target submarine cable includes:
  • the sensing equipment includes at least one of the following: ship monitoring equipment, optical fiber disturbance monitoring equipment, temperature detection equipment, and strain monitoring equipment;
  • the sensing device may exist in the following system:
  • a ship monitoring system that integrates radar photoelectric integration and AIS (Automatic identification System, ship automatic identification system).
  • the maritime radar detection system is a set of solid-state active phased radar system. During the working process, the system can search for the sea surface and low-altitude targets beyond the line of sight, find the target entering the search range, track the target, and establish the navigation trajectory of the target. It has the function of measuring motion attributes such as target heading and speed.
  • the radar antenna When the radar antenna is working, it periodically transmits ultra-high frequency electromagnetic wave signals to the surroundings, and at the same time receives the reflected electromagnetic waves in real time, and uses a set of calculation models to determine the size, distance and other information of the occluded object, so as to detect the object. effect.
  • the radar signal After the radar signal is sent out, it is reflected back when it encounters the object to be measured, and the reflected signal provides orientation information, which is converted into a digital quantity by the A/D conversion module, and then polar coordinates are converted.
  • the technology forms usable radar charts.
  • the vibration of the submarine cable will cause the optical fiber to deform, causing the length of the optical fiber and the refractive index of the optical fiber to change, which in turn makes the phase of the optical fiber.
  • the change is processed by the optical system, and the weak phase change is converted into a light intensity change. After photoelectric conversion and signal processing, it enters the computer for data analysis. Based on the analysis results, the system judges whether the disturbance/vibration submarine cable endangers the safe operation of the submarine cable. When the set value is exceeded, it will issue early warning and alarm signals in a timely manner, and confirm the vibration disturbance fault point, providing real-time information for subsequent processing.
  • the DTS system is introduced into the submarine cable monitoring system, and combined with the BOTDR, the temperature and strain of the submarine cable can be independently monitored, as shown in Figure 3-4.
  • the fiber monitored by DTS and the fiber monitored by BOTDR are located in the same optical unit of the submarine cable, so the temperature of the two fibers can always be kept the same.
  • the temperature change of the fiber is measured by DTS
  • the Brillouin frequency shift of the fiber is measured by BOTDR
  • the strain information of the fiber can be solved by signal processing.
  • the temperature and strain are separated, so that the system can independently monitor the temperature and strain of the submarine cable, improve the accuracy of the monitoring of the submarine cable strain, and can realize the damage to ships such as submarine cable hooks and smashes.
  • the real-time monitoring of behavior can detect the occurrence of ship anchor damage in time and carry out fast and accurate positioning, and can also monitor the current carrying capacity of submarine cables through temperature monitoring.
  • the deployment data includes environmental information and/or cable orientation information of the target submarine cable
  • the sensing data includes at least one of the following:
  • the specifically obtained data includes:
  • Submarine cable operation data Using DCR cable dynamic current carrying capacity technology, through conductor temperature modeling, optical fiber temperature measurement, integrated temperature analysis, alarm value setting, forming a curve to achieve the highest temperature point and abnormal peak monitoring. Based on OTDR (optical time-domain reflectometer, optical time-domain reflectometer) distributed optical fiber vibration sensing technology, study the interaction law of optical fiber Brillouin scattering frequency shift, strength and submarine cable strain and temperature changes, and obtain the normal working status of submarine cables and the strain and temperature data under the fault state; the optical fiber strain and temperature data and the operation state of the submarine cable can accurately determine the fault.
  • OTDR optical time-domain reflectometer, optical time-domain reflectometer
  • Submarine geomorphological data is usually used to collect seabed topography data, and the obtained data can be used as protection analysis of riprap dams, analysis of cast iron casings of submarine cables, and other features on the seabed (such as fishing nets temporarily placed by fishermen, such as shrimp cages, nets, etc.). source of basis for analysis.
  • Hydrological data Usually, in order to obtain the hydrological environment of the submarine cable, the water flow velocity in the routing area of the submarine cable is selected to be measured, and the force analysis of the submarine cable or its protection facilities can be carried out on this basis.
  • the direction and crossing data of submarine cables are mainly used to detect magnetic anomalies with marine magnetometers, to accurately detect and locate pipelines, cables and obstacles under the mud, and the generated data is usually in the format of MAG. It is necessary to know the intersection of the designed submarine cable and other submarine pipelines (such as submarine oil pipelines, submarine optical cable communication lines). In order to prevent damage to other existing submarine pipelines during the construction of submarine cables.
  • Submarine cable buried depth data when there is a submarine cable on the seabed, because the material properties of the submarine cable and the properties of the medium around the submarine cable are very different, the shallow (middle) formation profiler is used in the physical detection work. Pass vertically above. After the professional geophysical detection staff has identified the submarine cable, if the buried depth information of the submarine cable needs to be obtained, it is necessary to find the parabola vertex, that is, the spatial position of the submarine cable. The difference between the elevation of the parabola vertex and the seabed elevation of the corresponding position is Burial depth of submarine cables.
  • Submarine cable exposed dangling data The detection method of the exposed and suspended submarine cable is consistent with the detection method of the buried depth of the submarine cable.
  • the detection result of the buried depth of the submarine cable when the obtained burial depth data is 0, it means that this section of the submarine cable is in a bare state, and when the buried depth of the submarine cable is detected.
  • the depth data When the depth data is negative, it means that this section of submarine cable is in a suspended state.
  • auxiliary terrain detection and ROV (Remotely Operated Unmanned Vehicle) camera can be used to visually determine whether the submarine cable is completely exposed, partially exposed or in a critical state of exposure.
  • the above data can also be combined with the digital twin to perform position fusion, and coordinate matching is performed on all submarine cable engineering data.
  • the engineering information of each submarine cable is organized with the name of the submarine cable as the main identification ID. Display the cable name, supply oil field name, voltage level, submarine cable manufacturer information, planning time, construction time, completion time, management unit, operation and maintenance team, submarine cable material and other information in the form of attribute table.
  • the method further includes:
  • an alarm message is sent to the target terminal device, so that the target terminal device displays the alarm message, wherein the alarm message includes at least one of the following: the identification of the target submarine cable, Fault type and location message.
  • the twin database can store the result of each fault diagnosis, and the diagnosis result can be used as a training sample of the fault diagnosis model, and can also be used to monitor the operation state of the target submarine cable.
  • the construction of shore-based power supply projects in offshore oil fields inevitably involves a large number of submarine cables. Due to the special nature of its deep burial in the seabed, a huge amount of money and time are needed to maintain the submarine cables after failure. Therefore, the monitoring work for the operation status and faults of the submarine cables is essential.
  • the main factors that affect the safety of submarine cables are: the factors of the submarine cable itself, anchor damage, temperature, and natural factors. Therefore, the introduction of safe and effective monitoring and protection measures has become a necessary means to maintain monitoring of its location, path, buried depth and damage degree, which can effectively reduce maintenance costs.
  • the results of the economic benefit analysis of the submarine cable monitoring scheme for offshore wind farms show that the submarine cable monitoring system can effectively achieve the purpose of reducing the maintenance cost of the submarine cable, increasing the current carrying capacity of the submarine cable to increase the power generation and other purposes of improving economic benefits.
  • the submarine cable comprehensive online monitoring system is designed with the characteristics of safe, reliable and economical operation.
  • the extensive application of shore-based power supply in offshore oilfields, data collection, comparative analysis, optimization of calculation methods, and application of buried depth technology have improved the integrity and reliability of the monitoring system and played an important role in promoting the development of the power industry.
  • the present invention first obtains the deployment data and sensing data of the target submarine cable; then input the deployment data and the sensing data into a trained fault diagnosis model to obtain the diagnosis result of the target submarine cable, wherein , the fault diagnosis model is a probabilistic neural network model, and the diagnosis result includes the fault type.
  • the invention uses the trained probability neural network model as the fault diagnosis model to diagnose the fault of the target submarine cable, and can quickly and accurately obtain the fault type of the target submarine cable.
  • FIG. 2 shows a schematic structural diagram of a submarine cable fault diagnosis device provided by an embodiment of the present invention. For the convenience of description, only the part related to the embodiment of the present invention is shown, and the details are as follows:
  • the submarine cable fault diagnosis device 2 includes:
  • an acquisition module 21 for acquiring deployment data and sensing data of the target submarine cable
  • the diagnosis module 22 is configured to input deployment data and sensing data into a trained fault diagnosis model to obtain a diagnosis result of the target submarine cable, wherein the fault diagnosis model is a probabilistic neural network model, and the diagnosis result includes the fault type.
  • the device further includes:
  • a model building module for building an initial fault diagnosis model before inputting deployment data and sensing data into the trained fault diagnosis model
  • the model training module is used to train the initial fault diagnosis model based on the differential evolution algorithm and training samples to obtain the trained fault diagnosis model.
  • model training module includes:
  • the smoothing parameter determination unit is used to optimize the smoothing parameters in the initial fault diagnosis model through the differential evolution algorithm to obtain the target smoothing parameter, wherein the item to be optimized in the differential evolution algorithm is the smoothing parameter, and the objective function is the fault diagnosis model pair.
  • the sample training unit is used to train the fault diagnosis model using the target smoothing parameter based on the training samples, so as to obtain the trained fault diagnosis model.
  • the smoothing parameter determination unit is also used to:
  • the optimal individual in the offspring population is determined as the target smoothing parameter; otherwise, the offspring population is taken as the new parent population, and jumps to "In the parent population" Randomly select a specified number of individuals as mutation objects, and perform evolutionary operations on the parent population based on the mutation objects and the objective function".
  • model training module further includes:
  • the sample acquisition unit is used to obtain training samples from the twin database before training the initial fault diagnosis model based on the differential evolution algorithm and the training samples to obtain the trained fault diagnosis model, wherein the twin database includes historical faults
  • the deployment data, sensing data, and fault types of submarine cables, and the training samples are generated based on the deployment data, sensing data, and fault types of historically faulty submarine cables.
  • the obtaining module 21 is also used for:
  • the digital twin is updated.
  • the obtaining module 21 is also used for:
  • the sensing equipment includes at least one of the following: ship monitoring equipment, optical fiber disturbance monitoring equipment, temperature detection equipment and strain monitoring equipment;
  • the deployment data includes environmental information and/or cable orientation information of the target submarine cable, and the sensing data includes at least one of the following:
  • the diagnosis result is stored in the twin database
  • an alarm message is sent to the target terminal device, so that the target terminal device displays the alarm message, wherein the alarm message includes at least one of the following: the identification of the target submarine cable, Fault type and location message.
  • the acquisition module 21 , the diagnosis module 22 , the model establishment module, the model training module, the smoothing parameter determination unit, the sample training unit, and the sample acquisition unit may respectively be one having a communication interface and can realize the communication protocol. or multiple processors or chips, if necessary, may also include a memory and related interfaces, a system transmission bus, etc.; the processors or chips execute program-related codes to achieve corresponding functions.
  • the acquisition module 21, the diagnosis module 22, the model establishment module, the model training module, the smoothing parameter determination unit, the sample training unit, and the sample acquisition unit share an integrated chip or share equipment such as processors and memories .
  • the shared processor or chip executes program-related codes to implement corresponding functions.
  • the present invention first obtains the deployment data and sensing data of the target submarine cable; then input the deployment data and the sensing data into a trained fault diagnosis model to obtain the diagnosis result of the target submarine cable, wherein , the fault diagnosis model is a probabilistic neural network model, and the diagnosis result includes the fault type.
  • the invention uses the trained probability neural network model as the fault diagnosis model to diagnose the fault of the target submarine cable, and can quickly and accurately obtain the fault type of the target submarine cable.
  • FIG. 3 is a schematic diagram of an electronic device provided by an embodiment of the present invention.
  • the electronic device 3 of this embodiment includes: a processor 30 , a memory 31 , and a computer program 32 stored in the memory 31 and executable on the processor 30 .
  • the processor 30 executes the computer program 32
  • the steps in each of the above-mentioned embodiments of the submarine cable fault diagnosis method are implemented, for example, steps 101 to 102 shown in FIG. 1 .
  • the processor 30 executes the computer program 32
  • the functions of the modules/units in the above device embodiments, such as the functions of the modules 21 to 22 shown in FIG. 2 are implemented.
  • the computer program 32 can be divided into one or more modules/units, and the one or more modules/units are stored in the memory 31 and executed by the processor 30 to complete the this invention.
  • the one or more modules/units may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program 32 in the electronic device 3 .
  • the electronic device 3 may be a computing device such as a desktop computer, a notebook, a palmtop computer, and a cloud server.
  • the electronic device may include, but is not limited to, the processor 30 and the memory 31 .
  • FIG. 3 is only an example of the electronic device 3 , and does not constitute a limitation on the electronic device 3 , and may include more or less components than those shown in the figure, or combine some components, or different components
  • the electronic device may further include an input and output device, a network access device, a bus, and the like.
  • the so-called processor 30 may be a central processing unit (Central Processing Unit, CPU), or other general-purpose processors, digital signal processors (Digital Signal Processors, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • a general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
  • the memory 31 may be an internal storage unit of the electronic device 3 , such as a hard disk or a memory of the electronic device 3 .
  • the memory 31 may also be an external storage device of the electronic device 3, such as a plug-in hard disk equipped on the electronic device 3, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) Card, Flash Card, etc.
  • the memory 31 may also include both an internal storage unit of the electronic device 3 and an external storage device.
  • the memory 31 is used to store the computer program and other programs and data required by the electronic device.
  • the memory 31 can also be used to temporarily store data that has been output or will be output.
  • the disclosed apparatus/terminal and method may be implemented in other manners.
  • the apparatus/equipment embodiments described above are only illustrative.
  • the division of the modules or units is only a logical function division.
  • the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this embodiment.
  • each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit.
  • the above-mentioned integrated units may be implemented in the form of hardware, or may be implemented in the form of software functional units.
  • the integrated modules/units if implemented in the form of software functional units and sold or used as independent products, may be stored in a computer-readable storage medium.
  • the present invention can implement all or part of the processes in the methods of the above embodiments, and can also be completed by instructing relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium.
  • the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file or some intermediate form, and the like.
  • the computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, removable hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory) , Random Access Memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc.
  • ROM Read-Only Memory
  • RAM Random Access Memory
  • electric carrier signal telecommunication signal and software distribution medium, etc.
  • the content contained in the computer-readable media may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer-readable media Excluded are electrical carrier signals and telecommunication signals.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Mathematical Physics (AREA)
  • Biophysics (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Biomedical Technology (AREA)
  • Computing Systems (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Software Systems (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Probability & Statistics with Applications (AREA)
  • Physiology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Genetics & Genomics (AREA)
  • Testing Or Calibration Of Command Recording Devices (AREA)

Abstract

本发明适用于海底电缆维护技术领域，提供了一种海底电缆故障诊断方法、装置及设备，包括：获取目标海底电缆的部署数据和感测数据；将所述部署数据和所述感测数据输入经过训练的故障诊断模型，得到所述目标海底电缆的诊断结果，其中，所述故障诊断模型为概率神经网络模型，所述诊断结果包括故障类型。本发明将经过训练的概率神经网络模型作为故障诊断模型，对目标海底电缆进行故障诊断，可以快速准确的得到目标海底电缆的故障类型。

Description

一种海底电缆故障诊断方法、装置及设备 技术领域
本发明属于海底电缆维护技术领域，尤其涉及一种海底电缆故障诊断方法、装置及设备。
背景技术
目前，随着我国社会经济快速发展和海上油田等海洋开发利用持续深化，跨海输电网络不断以更高电压等级向沿海及其附近岛屿延伸覆盖，海底电缆应用越来越广泛。由于海洋环境恶劣，海底电缆在运行中受到海水长期侵蚀、海上作业船舶锚害等影响，故障时有发生，严重影响了跨海电网安全稳定运行。
然而，目前并没有技术可以实现对海底电缆进行实时监测，及时确定故障类型。
发明内容
有鉴于此，本发明提供了一种海底电缆故障诊断方法、装置及设备，以解决监测海底电缆状态，及时确定故障类型的问题。
本发明实施例的第一方面提供了一种海底电缆故障诊断方法，包括：
获取目标海底电缆的部署数据和感测数据；
将部署数据和感测数据输入经过训练的故障诊断模型，得到目标海底电缆的诊断结果，其中，故障诊断模型为概率神经网络模型，诊断结果包括故障类型。
可选的，在将部署数据和感测数据输入经过训练的故障诊断模型之前还包括：
建立初始的故障诊断模型；
基于差分进化算法和训练样本对初始的故障诊断模型进行训练，以得到经过训练的故障诊断模型。
可选的，基于差分进化算法和训练样本对初始的故障诊断模型进行训练，以得到经过训练的故障诊断模型包括：
通过差分进化算法对初始的故障诊断模型中的平滑参数进行优化，以得到目标平滑参数，其中，差分进化算法的待优化项为平滑参数，目标函数为故障诊断模型对训练样本进行故障诊断得到的故障类型与训练样本的真实故障类型的均方差，差分进化算法的最优解使目 标函数最小；
基于训练样本对采用目标平滑参数的故障诊断模型进行训练，以得到进过训练的故障诊断模型。
可选的，通过差分进化算法对初始的故障诊断模型中的平滑参数进行优化，以得到目标平滑参数包括：
在预设维度的空间生成预设数量且符合预设的约束条件的个体，得到初始种群，其中，每个个体分别为一个候选平滑参数；
将初始种群作为父代种群，在父代种群中随机选择指定个数的个体作为变异对象，基于变异对象和目标函数对父代种群进行进化操作，以得到子代种群；
若进化操作的总次数不小于预设值，则将子代种群中的最优个体确定为目标平滑参数；否则，将子代种群作为新的父代种群，并跳转至“在父代种群中随机选择指定个数的个体作为变异对象，基于变异对象和目标函数对父代种群进行进化操作”的步骤。
可选的，在基于差分进化算法和训练样本对初始的故障诊断模型进行训练，以得到经过训练的故障诊断模型之前，还包括：
从孪生数据库中获取训练样本，其中，孪生数据库中包括历史发生故障的海底电缆的部署数据、感测数据和故障类型，训练样本基于历史发生故障的海底电缆的部署数据、感测数据和故障类型生成。
可选的，方法还包括：
从工程数据库获取指定海底电缆的部署数据，从感测设备获取指定海底电缆的感测数据；
基于指定海底电缆的部署数据和感测数据，建立指定海底电缆的数字孪生体，其中，数字孪生体包括孪生数据库；
基于感测设备实时采集的感测数据，对数字孪生体进行更新。
可选的，获取目标海底电缆的部署数据和感测数据包括：
从孪生数据库获取目标海底电缆的部署数据和感测数据；
或，从工程数据库获取目标海底电缆的部署数据，从感测设备获取目标海底电缆的感测数据；
其中，感测设备包括下述至少一项：船舶监测设备、光纤扰动监测设备、温度检测设备和应变监测设备；
部署数据包括目标海底电缆的环境信息和/或电缆走向信息，感测数据包括下述至少一项：
目标海底电缆所在区域的船舶信息、扰动信息、温度信息和应变信息；
在将部署数据和感测数据输入经过训练的故障诊断模型，得到目标海底电缆的诊断结果之后，方法还包括：
将诊断结果存储至孪生数据库；
和/或，在目标海底电缆的诊断结果为发生故障时，向目标终端设备发生告警消息，以使目标终端设备显示告警消息，其中，告警消息包括下述至少一项：目标海底电缆的标识、故障类型和位置消息。
本发明实施例的第二方面提供了一种海底电缆故障诊断装置，包括：
获取模块，用于获取目标海底电缆的部署数据和感测数据；
诊断模块，用于将部署数据和感测数据输入经过训练的故障诊断模型，得到目标海底电缆的诊断结果，其中，故障诊断模型为概率神经网络模型，诊断结果包括故障类型。
可选的，该装置还包括：
模型建立模块，用于在将部署数据和感测数据输入经过训练的故障诊断模型之前，建立初始的故障诊断模型；
模型训练模块，用于基于差分进化算法和训练样本对初始的故障诊断模型进行训练，以得到经过训练的故障诊断模型。
可选的，模型训练模块包括：
平滑参数确定单元，用于通过差分进化算法对初始的故障诊断模型中的平滑参数进行优化，以得到目标平滑参数，其中，差分进化算法的待优化项为平滑参数，目标函数为故障诊断模型对训练样本进行故障诊断得到的故障类型与训练样本的真实故障类型的均方差，差分进化算法的最优解使目标函数最小；
样本训练单元，用于基于训练样本对采用目标平滑参数的故障诊断模型进行训练，以得到进过训练的故障诊断模型。
可选的，平滑参数确定单元还用于：
在预设维度的空间生成预设数量且符合预设的约束条件的个体，得到初始种群，其中，每个个体分别为一个候选平滑参数；
将初始种群作为父代种群，在父代种群中随机选择指定个数的个体作为变异对象，基于变异对象和目标函数对父代种群进行进化操作，以得到子代种群；
若进化操作的总次数不小于预设值，则将子代种群中的最优个体确定为目标平滑参数；否则，将子代种群作为新的父代种群，并跳转至“在父代种群中随机选择指定个数的个体作为变异对象，基于变异对象和目标函数对父代种群进行进化操作”的步骤。
可选的，模型训练模块还包括：
样本获取单元，用于在基于差分进化算法和训练样本对初始的故障诊断模型进行训练，以得到经过训练的故障诊断模型之前，从孪生数据库中获取训练样本，其中，孪生数据库中包括历史发生故障的海底电缆的部署数据、感测数据和故障类型，训练样本基于历史发生故障的海底电缆的部署数据、感测数据和故障类型生成。
可选的，获取模块还用于：
从工程数据库获取指定海底电缆的部署数据，从感测设备获取指定海底电缆的感测数据；
基于指定海底电缆的部署数据和感测数据，建立指定海底电缆的数字孪生体，其中，数字孪生体包括孪生数据库；
基于感测设备实时采集的感测数据，对数字孪生体进行更新。
可选的，获取模块还用于：
从孪生数据库获取目标海底电缆的部署数据和感测数据；
或，从工程数据库获取目标海底电缆的部署数据，从感测设备获取目标海底电缆的感测数据；
其中，感测设备包括下述至少一项：船舶监测设备、光纤扰动监测设备、温度检测设备和应变监测设备；
部署数据包括目标海底电缆的环境信息和/或电缆走向信息，感测数据包括下述至少一项：
目标海底电缆所在区域的船舶信息、扰动信息、温度信息和应变信息；
在将部署数据和感测数据输入经过训练的故障诊断模型，得到目标海底电缆的诊断结果之后，将诊断结果存储至孪生数据库；
和/或，在目标海底电缆的诊断结果为发生故障时，向目标终端设备发生告警消息，以使目标终端设备显示告警消息，其中，告警消息包括下述至少一项：目标海底电缆的标识、故障类型和位置消息。
本发明实施例的第三方面提供了一种电子设备，包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现如任一项所述海底电缆故障诊断方法的步骤。
本发明实施例的第四方面提供了一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行时实现如任一项所述海底电缆故障诊断方法的步骤。
本发明与现有技术相比存在的有益效果是：
本发明提供了一种海底电缆故障诊断方法，包括：获取目标海底电缆的部署数据和感测数据；将所述部署数据和所述感测数据输入经过训练的故障诊断模型，得到所述目标海底电缆的诊断结果，其中，所述故障诊断模型为概率神经网络模型，所述诊断结果包括故障类型。本发明通过经过训练的概率神经网络模型作为故障诊断模型，对目标海底电缆进行故障诊断，可以快速准确的得到目标海底电缆的故障类型。
附图说明
为了更清楚地说明本发明实施例中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。
图1是本发明实施例提供的海底电缆故障诊断方法的实现流程图；
图2是本发明实施例提供的海底电缆故障诊断装置的结构示意图；
图3是本发明实施例提供的电子设备的示意图。
具体实施方式
以下描述中，为了说明而不是为了限定，提出了诸如特定***结构、技术之类的具体细节，以便透彻理解本发明实施例。然而，本领域的技术人员应当清楚，在没有这些具体细节的其它实施例中也可以实现本发明。在其它情况中，省略对众所周知的***、装置、电路以及方法的详细说明，以免不必要的细节妨碍本发明的描述。
为使本发明的目的、技术方案和优点更加清楚，下面将结合附图通过具体实施例来进行说明。
参见图1，其示出了本发明实施例提供的海底电缆故障诊断方法的实现流程图，详述如下：
步骤101，获取目标海底电缆的部署数据和感测数据；
在本实施例中，可以根据指定电缆位置和电缆数量确定目标海底电缆。
步骤102，将部署数据和感测数据输入经过训练的故障诊断模型，得到目标海底电缆的诊断结果，其中，故障诊断模型为概率神经网络模型，诊断结果包括故障类型。
在本实施例中，采用概率神经网络模型作为故障诊断模型，概率神经网络是一种前馈 型神经网络，其具有简单的结构、网络训练过程快速的优点。海底电缆故障诊断模型中的概率神经网络由输入层、模式层、求和层和输出层共4层组成，能够对数据进行实时处理。
可选的，在将部署数据和感测数据输入经过训练的故障诊断模型之前还包括：
建立初始的故障诊断模型；
在本实施例中，海底电缆故障诊断模型中的概率神经网络应用的概率密度函数估计式如下:
Figure PCTCN2021113163-appb-000001
其中，X为要判断的故障样本，X ai为故障模式的第i个训练向量；P为故障数据样本的维数；m为故障模式的训练样本数目，δ为平滑参数，其取值对分类结果起着关键的作用，通常为人为凭经验给定。
基于差分进化算法和训练样本对初始的故障诊断模型进行训练，以得到经过训练的故障诊断模型。
可选的，基于差分进化算法和训练样本对初始的故障诊断模型进行训练，以得到经过训练的故障诊断模型包括下述步骤：
步骤一、通过差分进化算法对初始的故障诊断模型中的平滑参数进行优化，以得到目标平滑参数，其中，差分进化算法的待优化项为平滑参数，目标函数为故障诊断模型对训练样本进行故障诊断得到的故障类型与训练样本的真实故障类型的均方差，差分进化算法的最优解使目标函数最小；
在本实施例中，通过差分进化算法优化初始的故障诊断模型中的平滑参数，以提高故障诊断模型的分类速度和精度。
可选的，对初始的故障诊断模型中的平滑参数进行优化，以得到目标平滑参数包括：
在预设维度的空间生成预设数量且符合预设的约束条件的个体，得到初始种群，其中，每个个体分别为一个候选平滑参数；
将初始种群作为父代种群，在父代种群中随机选择指定个数的个体作为变异对象，基于变异对象和目标函数对父代种群进行进化操作，以得到子代种群；
若进化操作的总次数不小于预设值，则将子代种群中的最优个体确定为目标平滑参数；否则，将子代种群作为新的父代种群，并跳转至“在父代种群中随机选择指定个数的个体作为变异对象，基于变异对象和目标函数对父代种群进行进化操作”的步骤。
在本实施例中，使用了差分进化算法对平滑参数进行优化，差分进化算法是一种随机 启发式搜索算法，优点在于能够进行全局性的寻优，具有精度高、收敛速度快的优点。本实施例中使用的运行参数包括缩放因子F、交叉因子CR、群体规模M和最大进化代数G，进化操作由变异、交叉和选择三个过程组成，具体包括以下步骤：
(1)生成初始种群
在n维空间中随机产生符合种群约束条件的M个个体，实施操作如下:
Figure PCTCN2021113163-appb-000002
其中，
Figure PCTCN2021113163-appb-000003
和
Figure PCTCN2021113163-appb-000004
分别是第j个个体的边界条件，即为第i类里第j个个体的最大最小值；rand ij(0,1)是[0，1]之间产生的随机小数；x ij(0)是初始数据集。
(2)变异操作
从种群中随机选择3个个体x P1、x P2、x P3，i≠p 1≠p 2≠p 3，则变异操作为:
h ij(t+1)＝x b1j(t)+F(x P2j(t)-x P3j(t))  (3)
其中,x P2j(t)-x P3j(t)为差异化向量，p 1、p 2、p 3为个体在种群中的序号，x b1j(t)为当前代中种群中最好的个体；h ij(t+1)是指基于第t次迭代求出的最优个体为第t+1次迭代生成一个个体。
(3)交叉操作
交叉操作的目的是增加种群的多样性，操作为：
Figure PCTCN2021113163-appb-000005
其中，randl ij为[0，1]之间的随机小数，CR为交叉概率，CR的取值范围为[0，1]；v ij(t+1)是指第t+1次的迭代过程中生成一个个体，通过交叉操作将原有的x ij(t)进行替换生成一个新的样本，x ij(t)是指第t次迭代过程中第i类别第j元素。
(4)选择操作
选择操作用于确定目标向量x i1(t)能否成为种群的下一代个体，将目标向量x i1(t)与试验向量v i1(t+1)对应的目标函数值进行比较，并保留目标函数值更小的个体：
Figure PCTCN2021113163-appb-000006
其中，f(v i1(t+1))表示v i1(t+1)对应的目标函数值，f(x i1(t))表示x i1(t)对应的目标函数值；v i(t+1)表示对步骤(3)中v ij(t+1)进行操作，即对v i中的每个元素进行替换生成的个体。
重复执行步骤(2)～步骤(4)的操作，达到最大的进化代数G时停止进化，从最终种群中筛选出最优个体，即为平滑参数的最优解。本实施例中使用的差分进化算法可以快速的寻找平滑参数的最优解，也可以提高最优解的准确度，从而提高对海底电缆故障诊断的速度和准确度。
步骤二、基于训练样本对采用目标平滑参数的故障诊断模型进行训练，以得到进过训练的故障诊断模型。
可选的，本实施例还包括：
从工程数据库获取指定海底电缆的部署数据，从感测设备获取指定海底电缆的感测数据；
在本实施例中，指定海底电缆可以是指定区域内的海底电缆，可以与进行故障诊断的目标海底电缆相同，也可以不同；工程数据库包括海底电缆在进行施工时的数据，例如海底电缆所处环境的地形、水深、地貌、地质情况、水文数据，以及海底电缆的走向及交越数据。
基于指定海底电缆的部署数据和感测数据，建立指定海底电缆的数字孪生体，其中，数字孪生体包括孪生数据库；
基于感测设备实时采集的感测数据，对数字孪生体进行更新。
在本实施例中，物理实体由实际海底电缆和传感器***构成，传感器实时采集海底电缆运行数据，并将其传输至孪生数据库中进行储存，数字孪生体是物理实体在虚拟空间中的数字化映射，对海底电缆实体运行状态进行实时监测，其仿真模拟数据实时上传至数据库；各部分之间的连接可以实现数据的动态流动，数据的实时更新迭代，优化仿真的数字孪生体，使其更加接近海底电缆实体。通过对海底电缆故障诊断结果的分析，可以优化海底电缆检修维护方案，降低故障发生概率。
建立指定海底电缆的数字孪生体就是在虚拟空间建立指定海底电缆的高仿真度的模型，指定海底电缆的物理实体在虚拟空间的数字化描述，其利用数字模型、可靠感知，将数据实时输送至信息层，实现智能感知、实时交互。本实施例中的数字孪生体集成与融合了几何、物理、行为及规则4层模型，能够实时采集故障数据，模拟指定海底电缆的实际运行状态，在多维度、多时间尺度上对物理实体进行高保真度地描述，建立实时映射，可以实现模拟、监控、诊断物理实体在现实环境中的状态和行为。基于感测设备实时采集的感测数据，对数 字孪生体进行更新，可以使数字孪生体更接近指定海底电缆实体。
指定海底电缆数字孪生体的构建，提供了一种数据驱动的方式全方位刻画描述设备在不同环境下老化过程、人的运维行为作用、以及与环境交互等复杂不确定性的演变过程，有助于实现海底电缆的全生命周期管理，有效降低海底电缆运行成本，提高海底电缆使用寿命。
借助数字孪生提供的全息数字镜像与可视化能力，常规手段难以获取的复杂环节内部状态信息能够通过虚拟现实、增强现实等技术手段直观呈现给运维人员，这给综合能源***的快速异常发现、故障检测与诊断提供了重要手段，现场人员可以基于数字孪生提供的信息快速定位故障点并实施修复；
利用数字孪生的多源数据集成利用能力，可基于***运行历史、发展态势、设备状态、同类型设备故障统计等信息进行综合研究判断，预测供能环节运行寿命和故障概率，支持更加精准的预测性维护；并为各种运维方案提供灵活高效的虚拟测试与评估环境。
可选的，在基于差分进化算法和训练样本对初始的故障诊断模型进行训练，以得到经过训练的故障诊断模型之前，还包括：
从孪生数据库中获取训练样本，其中，孪生数据库中包括历史发生故障的海底电缆的部署数据、感测数据和故障类型，训练样本基于历史发生故障的海底电缆的部署数据、感测数据和故障类型生成。
在本实施例中，孪生数据库中储存海底电缆实体的制造、原理、各部件以及构成数字孪生体几何、物理、行为及规则模型的所有数据信息、实时数据，虚拟海底电缆的仿真数据，在孪生数据库中获取训练样本对故障诊断模型进行训练，使故障诊断模型可以得到更准确的诊断结果。
可选的，获取目标海底电缆的部署数据和感测数据包括：
从孪生数据库获取目标海底电缆的部署数据和感测数据；
或，从工程数据库获取目标海底电缆的部署数据，从感测设备获取目标海底电缆的感测数据；
可选的，感测设备包括下述至少一项：船舶监测设备、光纤扰动监测设备、温度检测设备和应变监测设备；
在本实施例中，感测设备可以存在于下述***中：
(1)雷达光电一体化与AIS(Automatic identification System，船舶自动识别***)相融合的船舶监测***
海事雷达探测***是一套固态有源相控雷达***，***在工作过程中，可对视距外的海面和低空目标进行搜索，发现进入搜索范围的目标、跟踪目标、建立目标的航行轨迹，具 有对目标航向、航速等运动属性的测量功能。雷达天线在工作时，通过周期性的向周围发射超高频电磁波信号，同时实时接收反射回来的电磁波，通过一套计算模型来判断所遮挡物体的大小、距离等信息，从而起到探测物体的作用。
雷达信号发出后遇到被测物体后反射回来，反射信号提供方位信息，经A/D转换模块转换成数字量，再进行极坐标转换，后经图像处理后与海图坐标匹配，最后通过叠加技术形成可用雷达海图。
(2)海底电缆扰动监测
当海底电缆所在海域通过的船舶产生振动、或船舶锚挂扰动作用于海底电缆时，海底电缆的振动会导致光纤发生形变，引起了光纤长度以及光纤的折射率发生了变化，进而使得光纤的相位变化，经光学***处理，将微弱的相位变化转换为光强变化，经光电转换和信号处理后，进入计算机进行数据分析。***根据分析的结果，判断扰动/振动海底电缆电缆是否危及海底电缆安全运行，当超过设定值时，适时发出预警、报警信号，并确认振动扰动故障点，为后续处理提供实时信息。
(3)基于DTS(Distributed Temperature Sensing，分布式光纤测温***)和BOTDR(布里渊光时域反射技术)相融合的温度与应变独立监测***
在海底电缆监测***中引入DTS***，与BOTDR相结合，可进行海底电缆温度与应变独立监测，如图3－4所示。DTS所监测光纤与BOTDR所监测光纤位于海底电缆同一光单元中，因此两根光纤温度能始终保持一致。通过DTS测量出光纤温度的变化、BOTDR测量出光纤布里渊频移的变化，通过信号处理即可解出光纤的应变信息。通过DTS与BOTDR二者的融合，将温度与应变分离，使***可进行海底电缆温度与应变的独立监测，提高了对海底电缆应变监测的精度，可实现对海底电缆钩、砸等船舶锚害行为的实时监测，可及时发现船舶锚害情况发生并进行快速准确定位，也可通过温度监测获得对海底电缆载流量的监测。
相应的，部署数据包括目标海底电缆的环境信息和/或电缆走向信息，感测数据包括下述至少一项：
目标海底电缆所在区域的船舶信息、扰动信息、温度信息和应变信息；
在本实施例中，具体获取的数据包括：
(1)海底电缆运行数据。采用DCR电缆动态载流量技术，通过导体温度建模，光纤测温，集成温度分析、报警值设定，形成曲线，实现最高温度点和异常尖峰监测。基于OTDR(optical time-domain reflectometer，光时域反射仪)分布式光纤振动传感技术，研究光纤布里渊散射频移、强度与海底电缆应变、温度变化的作用规律，获取海底电缆正常工作状态和故障状态下的应变与温度数据；光纤应变和温度数据与海底电缆运行状态，能够准确的判定 故障。
(2)海底地形水深数据。三维海底场景的构建是海底电缆工程数据三维可视化的基础。海底地形测量通常采用单波束测深仪和多波束测深仪，对海底电缆区域进行水深测量，获得精密水下三维地形图。在浅水区多波束无法覆盖的地方，可采用数字双频测深仪采集水深数据，形成高精度水下地形图。
(3)海底地貌数据。海底地貌数据采集通常采用侧扫声呐技术手段，获取的数据可以作为抛石坝保护分析、海底电缆铸铁套管分析、海底其他地物(如渔民临时所放捕鱼网具，如虾笼、网绳、浮标等)分析的依据来源。
(4)海底地质数据。利用声波在不同的介质中其传播速度显著不同的原理，由于海底不同分界面内岩层的密度不同，那么声波在通过不同分界面时的反射系数也截然不同。从而在接收到的反射信号中携带了海底介质的地质信息，通过专业数据处理，最后输出的数据成果可直观的了解海底沉积物的属性。
(5)水文数据。通常为了获取海底电缆的水文环境，会选择测海底电缆路由区水流速度，并可在此基础上进行海底电缆或其保护设施受力分析。
(6)海底电缆走向及交越数据。海底电缆走向和交越点数据主要是利用海洋磁力仪进行磁异常探测，准确对管道、电缆及泥下障碍物进行探测与定位，而产生的数据，通常格式为MAG。需要了解设计的海底电缆与其他海底管道(如海底石油输送管道、海底光缆通信线)的交叉位置。以预防海底电缆施工期间对其他已建海底管道线路的破坏。
(7)海底电缆埋深数据。利用浅(中)地层剖面仪，当海底存在海底电缆时，因为海底电缆的材料性质和海底电缆周围介质的性质差异巨大，所以当进行物理探测工作时浅(中)地层剖面仪在海底电缆的上方垂直经过。待专业物探检测工作人员识别出海底电缆之后，如果还需获取海底电缆的埋深信息，需找到抛物线顶点，即海底电缆所处的空间位置，此抛物线顶点高程与相应位置海底高程之差即为海底电缆的埋深。
(8)海底电缆裸露悬空数据。海底电缆裸露悬空的检测方法与海底电缆埋深检测方法一致，在海底电缆埋深检测结果中，当获取的埋深数据为0时，代表该段海底电缆处于裸露状态，而当海底电缆的埋深数据为负值时，代表该段海底电缆处于悬空状态。此外，辅助地形地貌检测和ROV(遥控无人潜水器)摄像，以直观的方式判别海底电缆是完全裸露、部分裸露还是处于裸露临界状态。
在本实施例中，还可以将上述数据结合数字孪生体进行位置融合，对所有的海底电缆工程数据进行坐标匹配。在不同源数据集中找到表示同一地理实体的对象并抽取出来，并对其增加坐标这一属性。每一条海底电缆的工程信息，以海底电缆的名字为主要识别ID进行 命名组织。以属性表的形式展示其电缆名称、供给油田名称、电压等级、海底电缆厂家信息、规划时间、建设时间、竣工时间、管理单位、运维班组、海底电缆材质等信息。
可选的，在将部署数据和感测数据输入经过训练的故障诊断模型，得到目标海底电缆的诊断结果之后，还包括：
将诊断结果存储至孪生数据库；
和/或，在目标海底电缆的诊断结果为发生故障时，向目标终端设备发生告警消息，以使目标终端设备显示告警消息，其中，告警消息包括下述至少一项：目标海底电缆的标识、故障类型和位置消息。
在本实施例中，孪生数据库可以存储每次故障诊断的结果，诊断结果可以作为故障诊断模型的训练样本，也可以用于对目标海底电缆的运行状态进行监测。海上油田岸基供电项目的建设不可避免涉及大量海底电缆，由于其深埋海底的特殊性质，海底电缆故障后需要投入巨大的金钱与时间进行维修，因此，针对海底电缆运行状态与故障的监测工作是必不可少的。影响海底电缆安全的主要因素有：海底电缆自身因素、锚害、温度、自然因素。因此，引入安全有效的监测防护措施成为必要手段，保持对其位置、路径、埋深与损坏程度的监测，能够有效减少维护成本。
对海上风电场海底电缆监测方案经济效益分析的结果表明，海底电缆监测***能有效的实现降低海底电缆维护成本，提高海底电缆载流量以增加发电量等提高经济效益的目的。海底电缆综合在线监测***设计具有运行安全、可靠、经济等特性。在海上油田岸基供电的广泛运用，数据收集、比较分析，计算法的优化，埋深技术的应用，使得监控***完整性、可靠性得于提高，促进电力事业的发展有重要的作用。
由上可知，本发明首先获取目标海底电缆的部署数据和感测数据；然后将所述部署数据和所述感测数据输入经过训练的故障诊断模型，得到所述目标海底电缆的诊断结果，其中，所述故障诊断模型为概率神经网络模型，所述诊断结果包括故障类型。本发明通过经过训练的概率神经网络模型作为故障诊断模型，对目标海底电缆进行故障诊断，可以快速准确的得到目标海底电缆的故障类型。
图2示出了本发明实施例提供的海底电缆故障诊断装置的结构示意图，为了便于说明，仅示出了与本发明实施例相关的部分，详述如下：
如图2所示，海底电缆故障诊断装置2包括：
获取模块21，用于获取目标海底电缆的部署数据和感测数据；
诊断模块22，用于将部署数据和感测数据输入经过训练的故障诊断模型，得到目标海底电缆的诊断结果，其中，故障诊断模型为概率神经网络模型，诊断结果包括故障类型。
可选的，该装置还包括：
模型建立模块，用于在将部署数据和感测数据输入经过训练的故障诊断模型之前，建立初始的故障诊断模型；
模型训练模块，用于基于差分进化算法和训练样本对初始的故障诊断模型进行训练，以得到经过训练的故障诊断模型。
可选的，模型训练模块包括：
平滑参数确定单元，用于通过差分进化算法对初始的故障诊断模型中的平滑参数进行优化，以得到目标平滑参数，其中，差分进化算法的待优化项为平滑参数，目标函数为故障诊断模型对训练样本进行故障诊断得到的故障类型与训练样本的真实故障类型的均方差，差分进化算法的最优解使目标函数最小；
样本训练单元，用于基于训练样本对采用目标平滑参数的故障诊断模型进行训练，以得到进过训练的故障诊断模型。
可选的，平滑参数确定单元还用于：
在预设维度的空间生成预设数量且符合预设的约束条件的个体，得到初始种群，其中，每个个体分别为一个候选平滑参数；
将初始种群作为父代种群，在父代种群中随机选择指定个数的个体作为变异对象，基于变异对象和目标函数对父代种群进行进化操作，以得到子代种群；
若进化操作的总次数不小于预设值，则将子代种群中的最优个体确定为目标平滑参数；否则，将子代种群作为新的父代种群，并跳转至“在父代种群中随机选择指定个数的个体作为变异对象，基于变异对象和目标函数对父代种群进行进化操作”的步骤。
可选的，模型训练模块还包括：
样本获取单元，用于在基于差分进化算法和训练样本对初始的故障诊断模型进行训练，以得到经过训练的故障诊断模型之前，从孪生数据库中获取训练样本，其中，孪生数据库中包括历史发生故障的海底电缆的部署数据、感测数据和故障类型，训练样本基于历史发生故障的海底电缆的部署数据、感测数据和故障类型生成。
可选的，获取模块21还用于：
从工程数据库获取指定海底电缆的部署数据，从感测设备获取指定海底电缆的感测数据；
基于指定海底电缆的部署数据和感测数据，建立指定海底电缆的数字孪生体，其中，数字孪生体包括孪生数据库；
基于感测设备实时采集的感测数据，对数字孪生体进行更新。
可选的，获取模块21还用于：
从孪生数据库获取目标海底电缆的部署数据和感测数据；
或，从工程数据库获取目标海底电缆的部署数据，从感测设备获取目标海底电缆的感测数据；
其中，感测设备包括下述至少一项：船舶监测设备、光纤扰动监测设备、温度检测设备和应变监测设备；
部署数据包括目标海底电缆的环境信息和/或电缆走向信息，感测数据包括下述至少一项：
目标海底电缆所在区域的船舶信息、扰动信息、温度信息和应变信息；
在将部署数据和感测数据输入经过训练的故障诊断模型，得到目标海底电缆的诊断结果之后，将诊断结果存储至孪生数据库；
和/或，在目标海底电缆的诊断结果为发生故障时，向目标终端设备发生告警消息，以使目标终端设备显示告警消息，其中，告警消息包括下述至少一项：目标海底电缆的标识、故障类型和位置消息。
在本申请的实施例中，所述获取模块21，诊断模块22，模型建立模块、模型训练模块、平滑参数确定单元、样本训练单元、样本获取单元分别可以是具有通信接口能够实现通信协议的一个或多个处理器或者芯片，如有需要还可以包括存储器及相关的接口、***传输总线等；所述处理器或者芯片执行程序相关的代码实现相应的功能。或者，可替换的方案为，所述获取模块21，诊断模块22，模型建立模块、模型训练模块、平滑参数确定单元、样本训练单元、样本获取单元共享一个集成芯片或者共享处理器、存储器等设备。所述共享的处理器或者芯片执行程序相关的代码实现相应的功能。
由上可知，本发明首先获取目标海底电缆的部署数据和感测数据；然后将所述部署数据和所述感测数据输入经过训练的故障诊断模型，得到所述目标海底电缆的诊断结果，其中，所述故障诊断模型为概率神经网络模型，所述诊断结果包括故障类型。本发明通过经过训练的概率神经网络模型作为故障诊断模型，对目标海底电缆进行故障诊断，可以快速准确的得到目标海底电缆的故障类型。
图3是本发明一实施例提供的电子设备的示意图。如图3所示，该实施例的电子设备3包括：处理器30、存储器31以及存储在所述存储器31中并可在所述处理器30上运行的计算机程序32。所述处理器30执行所述计算机程序32时实现上述各个海底电缆故障诊断方法实施例中的步骤，例如图1所示的步骤101至步骤102。或者，所述处理器30执行所述计算机程序32时实现上述各装置实施例中各模块/单元的功能，例如图2所示模块21至22的功 能。
示例性的，所述计算机程序32可以被分割成一个或多个模块/单元，所述一个或者多个模块/单元被存储在所述存储器31中，并由所述处理器30执行，以完成本发明。所述一个或多个模块/单元可以是能够完成特定功能的一系列计算机程序指令段，该指令段用于描述所述计算机程序32在所述电子设备3中的执行过程。
所述电子设备3可以是桌上型计算机、笔记本、掌上电脑及云端服务器等计算设备。所述电子设备可包括，但不仅限于，处理器30、存储器31。本领域技术人员可以理解，图3仅仅是电子设备3的示例，并不构成对电子设备3的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件，例如所述电子设备还可以包括输入输出设备、网络接入设备、总线等。
所称处理器30可以是中央处理单元(Central Processing Unit，CPU)，还可以是其他通用处理器、数字信号处理器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现场可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
所述存储器31可以是所述电子设备3的内部存储单元，例如电子设备3的硬盘或内存。所述存储器31也可以是所述电子设备3的外部存储设备，例如所述电子设备3上配备的插接式硬盘，智能存储卡(Smart Media Card，SMC)，安全数字(Secure Digital，SD)卡，闪存卡(Flash Card)等。进一步地，所述存储器31还可以既包括所述电子设备3的内部存储单元也包括外部存储设备。所述存储器31用于存储所述计算机程序以及所述电子设备所需的其他程序和数据。所述存储器31还可以用于暂时地存储已经输出或者将要输出的数据。
所属领域的技术人员可以清楚地了解到，为了描述的方便和简洁，仅以上述各功能单元、模块的划分进行举例说明，实际应用中，可以根据需要而将上述功能分配由不同的功能单元、模块完成，即将所述装置的内部结构划分成不同的功能单元或模块，以完成以上描述的全部或者部分功能。实施例中的各功能单元、模块可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中，上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。另外，各功能单元、模块的具体名称也只是为了便于相互区分，并不用于限制本申请的保护范围。上述***中单元、模块的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述或记载的部分，可以参见其它实施例的相关描述。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本发明的范围。
在本发明所提供的实施例中，应该理解到，所揭露的装置/终端和方法，可以通过其它的方式实现。例如，以上所描述的装置/设备实施例仅仅是示意性的，例如，所述模块或单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个***，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通讯连接可以是通过一些接口，装置或单元的间接耦合或通讯连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本发明各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
所述集成的模块/单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本发明实现上述实施例方法中的全部或部分流程，也可以通过计算机程序来指令相关的硬件来完成，所述的计算机程序可存储于一计算机可读存储介质中，该计算机程序在被处理器执行时，可实现上述各个方法实施例的步骤。其中，所述计算机程序包括计算机程序代码，所述计算机程序代码可以为源代码形式、对象代码形式、可执行文件或某些中间形式等。所述计算机可读介质可以包括：能够携带所述计算机程序代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、电载波信号、电信信号以及软件分发介质等。需要说明的是，所述计算机可读介质包含的内容可以根据司法管辖区内立法和专利实践的要求进行适当的增减，例如在某些司法管辖区，根据立法和专利实践，计算机可读介质不包括是电载波信号和电信信号。
以上所述实施例仅用以说明本发明的技术方案，而非对其限制；尽管参照前述实施例对本发明进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换， 并不使相应技术方案的本质脱离本发明各实施例技术方案的精神和范围，均应包含在本发明的保护范围之内。

Claims (10)

  1. 一种海底电缆故障诊断方法，其特征在于，包括：
    获取目标海底电缆的部署数据和感测数据；
    将所述部署数据和所述感测数据输入经过训练的故障诊断模型，得到所述目标海底电缆的诊断结果，其中，所述故障诊断模型为概率神经网络模型，所述诊断结果包括故障类型。
  2. 根据权利要求1所述的海底电缆故障诊断方法，其特征在于，在将所述部署数据和所述感测数据输入经过训练的故障诊断模型之前还包括：
    建立初始的故障诊断模型；
    基于差分进化算法和训练样本对所述初始的故障诊断模型进行训练，以得到经过训练的故障诊断模型。
  3. 根据权利要求2所述的海底电缆故障诊断方法，其特征在于，所述基于差分进化算法和训练样本对所述初始的故障诊断模型进行训练，以得到经过训练的故障诊断模型包括：
    通过差分进化算法对所述初始的故障诊断模型中的平滑参数进行优化，以得到目标平滑参数，其中，所述差分进化算法的待优化项为平滑参数，目标函数为故障诊断模型对训练样本进行故障诊断得到的故障类型与所述训练样本的真实故障类型的均方差，所述差分进化算法的最优解使所述目标函数最小；
    基于所述训练样本对采用所述目标平滑参数的故障诊断模型进行训练，以得到进过训练的故障诊断模型。
  4. 根据权利要求3所述的海底电缆故障诊断方法，其特征在于，所述通过差分进化算法对所述初始的故障诊断模型中的平滑参数进行优化，以得到目标平滑参数包括：
    在预设维度的空间生成预设数量且符合预设的约束条件的个体，得到初始种群，其中，每个个体分别为一个候选平滑参数；
    将所述初始种群作为父代种群，在所述父代种群中随机选择指定个数的个体作为变异对象，基于所述变异对象和所述目标函数对所述父代种群进行进化操作，以得到子代种群；
    若进化操作的总次数不小于预设值，则将所述子代种群中的最优个体确定为所述目标平滑参数；否则，将所述子代种群作为新的父代种群，并跳转至“在所述父代种群中随机选择指定个数的个体作为变异对象，基于所述变异对象和所述目标函数对所述父代种群进行进化操作”的步骤。
  5. 根据权利要求2所述的海底电缆故障诊断方法，其特征在于，在所述基于差分进化 算法和训练样本对所述初始的故障诊断模型进行训练，以得到经过训练的故障诊断模型之前，还包括：
    从孪生数据库中获取所述训练样本，其中，所述孪生数据库中包括历史发生故障的海底电缆的部署数据、感测数据和故障类型，所述训练样本基于所述历史发生故障的海底电缆的部署数据、感测数据和故障类型生成。
  6. 根据权利要求5所述的海底电缆故障诊断方法，其特征在于，所述方法还包括：
    从工程数据库获取指定海底电缆的部署数据，从感测设备获取指定海底电缆的感测数据；
    基于所述指定海底电缆的部署数据和感测数据，建立所述指定海底电缆的数字孪生体，其中，所述数字孪生体包括孪生数据库；
    基于感测设备实时采集的感测数据，对所述数字孪生体进行更新。
  7. 根据权利要求1至6任一项所述的海底电缆故障诊断方法，其特征在于，所述获取目标海底电缆的部署数据和感测数据包括：
    从孪生数据库获取所述目标海底电缆的部署数据和感测数据；
    或，从工程数据库获取所述目标海底电缆的部署数据，从感测设备获取所述目标海底电缆的感测数据；
    其中，感测设备包括下述至少一项：船舶监测设备、光纤扰动监测设备、温度检测设备和应变监测设备；
    所述部署数据包括所述目标海底电缆的环境信息和/或电缆走向信息，所述感测数据包括下述至少一项：
    所述目标海底电缆所在区域的船舶信息、扰动信息、温度信息和应变信息；
    在将所述部署数据和所述感测数据输入经过训练的故障诊断模型，得到所述目标海底电缆的诊断结果之后，所述方法还包括：
    将所述诊断结果存储至所述孪生数据库；
    和/或，在所述目标海底电缆的诊断结果为发生故障时，向目标终端设备发生告警消息，以使所述目标终端设备显示所述告警消息，其中，所述告警消息包括下述至少一项：目标海底电缆的标识、故障类型和位置消息。
  8. 一种海底电缆故障诊断装置，其特征在于，包括：
    获取模块，用于获取目标海底电缆的部署数据和感测数据；
    诊断模块，用于将所述部署数据和所述感测数据输入经过训练的故障诊断模型，得到所述目标海底电缆的诊断结果，其中，所述故障诊断模型为概率神经网络模型，所述诊断结 果包括故障类型。
  9. 一种电子设备，包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机程序，其特征在于，所述处理器执行所述计算机程序时实现如上的权利要求1至7中任一项所述海底电缆故障诊断方法的步骤。
  10. 一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，其特征在于，所述计算机程序被处理器执行时实现如上的权利要求1至7中任一项所述海底电缆故障诊断方法的步骤。
PCT/CN2021/113163 2021-03-24 2021-08-18 一种海底电缆故障诊断方法、装置及设备 WO2022198899A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/599,552 US11774485B2 (en) 2021-03-24 2021-08-18 Fault diagnosis method and apparatus for submarine cable, and device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110313246.4 2021-03-24
CN202110313246.4A CN113298110A (zh) 2021-03-24 2021-03-24 一种海底电缆故障诊断方法、装置及设备

Publications (1)

Publication Number Publication Date
WO2022198899A1 true WO2022198899A1 (zh) 2022-09-29

Family

ID=77319191

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/113163 WO2022198899A1 (zh) 2021-03-24 2021-08-18 一种海底电缆故障诊断方法、装置及设备

Country Status (3)

Country Link
US (1) US11774485B2 (zh)
CN (1) CN113298110A (zh)
WO (1) WO2022198899A1 (zh)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115333923A (zh) * 2022-10-14 2022-11-11 成都飞机工业（集团）有限责任公司 一种故障点溯源分析方法、装置、设备及介质
CN115388959A (zh) * 2022-10-31 2022-11-25 高勘(广州)技术有限公司 海底电缆运维方法、装置、设备及存储介质
CN116150676A (zh) * 2023-04-19 2023-05-23 山东能源数智云科技有限公司 基于人工智能的设备故障诊断与识别方法及装置
CN116520068A (zh) * 2023-07-04 2023-08-01 深圳博润缘科技有限公司 一种电力数据的诊断方法、装置、设备及存储介质
CN116626543A (zh) * 2023-07-21 2023-08-22 无锡市环球电器装备有限公司 基于集肤效应的加热电缆故障检测方法及***
CN116819345A (zh) * 2023-08-29 2023-09-29 武汉船用电力推进装置研究所(中国船舶集团有限公司第七一二研究所) 电池***故障识别方法、装置、电子设备及存储介质
CN117309624A (zh) * 2023-11-30 2023-12-29 天津天大求实电力新技术股份有限公司 一种感温电缆性能评价方法及***

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114166890B (zh) * 2021-11-05 2022-07-19 西南交通大学 一种车载epr电缆运行年龄的估算方法
CN114279676B (zh) * 2021-11-23 2024-06-07 国核电力规划设计研究院有限公司 海底悬空电缆应变损伤的海况阈值判断方法
CN115014571B (zh) * 2022-05-25 2023-03-24 汕头大学 一种海底电缆风险事件识别***及方法
CN115424365B (zh) * 2022-08-11 2023-11-24 广东联塑精铟科技有限公司 一种基于数字孪生的网衣实时监测方法及***
CN115293303B (zh) * 2022-10-10 2023-01-24 广东电网有限责任公司中山供电局 一种高压输电线网监测方法、***、设备和介质
CN116658489B (zh) * 2023-06-02 2023-12-01 上海电气液压气动有限公司 一种基于数字孪生的液压***故障诊断方法及***
CN116520270B (zh) * 2023-07-04 2023-09-05 四川天中星航空科技有限公司 一种基于评估模型的雷达电子战测试方法
CN117232577B (zh) * 2023-09-18 2024-04-05 杭州奥克光电设备有限公司 一种光缆交接箱承载内部监测方法、***和光缆交接箱
CN117176550B (zh) * 2023-09-25 2024-03-19 云念软件(广东)有限公司 基于故障辨识的集成运行维护方法及***
CN117011690B (zh) * 2023-10-07 2024-02-09 广东电网有限责任公司阳江供电局 一种海缆隐患识别方法、装置、设备和介质
CN117726145B (zh) * 2024-02-07 2024-05-10 广东电网有限责任公司广州供电局 电缆巡检处理方法、装置、电子设备和计算机可读介质
CN117974913B (zh) * 2024-04-02 2024-06-11 深圳供电局有限公司 配网电缆机器人控制方法、装置、电子设备和介质

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101819253A (zh) * 2010-04-20 2010-09-01 湖南大学 一种基于概率神经网络的容差电路故障诊断方法
CN111504635A (zh) * 2020-04-21 2020-08-07 哈尔滨理工大学 基于差分进化概率神经网络的行星齿轮故障诊断方法

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102981104B (zh) * 2012-11-19 2015-03-11 中国能源建设集团广东省电力设计研究院 海底电缆在线监测方法
US11637417B2 (en) * 2018-09-06 2023-04-25 City University Of Hong Kong System and method for analyzing survivability of an infrastructure link
CN111650921A (zh) * 2020-05-20 2020-09-11 国网江苏省电力有限公司泰州供电分公司 一种智能电网调控控制***设备故障诊断方法及***

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101819253A (zh) * 2010-04-20 2010-09-01 湖南大学 一种基于概率神经网络的容差电路故障诊断方法
CN111504635A (zh) * 2020-04-21 2020-08-07 哈尔滨理工大学 基于差分进化概率神经网络的行星齿轮故障诊断方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
LYU ANQIANG ，, LIU ZHENG;YIN CHENG-QUN;LI YONG-QIAN: "A Fault Diagnosis Method Forwavelet Packet and Neural Network-Based Submarine Cables", GUANGTONGXIN-YANJIU = STUDY ON OPTICAL COMMUNICATIONS, WUHAN, no. 2, 30 April 2016 (2016-04-30), pages 26 - 29, XP055970547, ISSN: 1005-8788, DOI: 10.13756/j.gtxyj.2016.02.009 *

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115333923A (zh) * 2022-10-14 2022-11-11 成都飞机工业（集团）有限责任公司 一种故障点溯源分析方法、装置、设备及介质
CN115333923B (zh) * 2022-10-14 2023-03-14 成都飞机工业（集团）有限责任公司 一种故障点溯源分析方法、装置、设备及介质
CN115388959A (zh) * 2022-10-31 2022-11-25 高勘(广州)技术有限公司 海底电缆运维方法、装置、设备及存储介质
CN116150676A (zh) * 2023-04-19 2023-05-23 山东能源数智云科技有限公司 基于人工智能的设备故障诊断与识别方法及装置
CN116150676B (zh) * 2023-04-19 2023-09-26 山东能源数智云科技有限公司 基于人工智能的设备故障诊断与识别方法及装置
CN116520068B (zh) * 2023-07-04 2023-09-22 深圳博润缘科技有限公司 一种电力数据的诊断方法、装置、设备及存储介质
CN116520068A (zh) * 2023-07-04 2023-08-01 深圳博润缘科技有限公司 一种电力数据的诊断方法、装置、设备及存储介质
CN116626543A (zh) * 2023-07-21 2023-08-22 无锡市环球电器装备有限公司 基于集肤效应的加热电缆故障检测方法及***
CN116626543B (zh) * 2023-07-21 2023-09-26 无锡市环球电器装备有限公司 基于集肤效应的加热电缆故障检测方法及***
CN116819345A (zh) * 2023-08-29 2023-09-29 武汉船用电力推进装置研究所(中国船舶集团有限公司第七一二研究所) 电池***故障识别方法、装置、电子设备及存储介质
CN116819345B (zh) * 2023-08-29 2023-11-21 武汉船用电力推进装置研究所(中国船舶集团有限公司第七一二研究所) 电池***故障识别方法、装置、电子设备及存储介质
CN117309624A (zh) * 2023-11-30 2023-12-29 天津天大求实电力新技术股份有限公司 一种感温电缆性能评价方法及***
CN117309624B (zh) * 2023-11-30 2024-02-06 天津天大求实电力新技术股份有限公司 一种感温电缆性能评价方法及***

Also Published As

Publication number Publication date
US11774485B2 (en) 2023-10-03
CN113298110A (zh) 2021-08-24
US20230160939A1 (en) 2023-05-25

Similar Documents

Publication Publication Date Title
WO2022198899A1 (zh) 一种海底电缆故障诊断方法、装置及设备
CN109243205B (zh) 一种沿海水上交通安全风险监测与预警***及方法
Fang et al. Automatic identification system-based approach for assessing the near-miss collision risk dynamics of ships in ports
JP5905646B2 (ja) 津波監視システム
Xia et al. Operation and maintenance optimization of offshore wind farms based on digital twin: A review
CN117408536A (zh) 基于ai的水下测绘实时分析***
CN116757097A (zh) 一种数字孪生水利工程运维监测***和方法
CN117217058A (zh) 一种抽蓄电站水工结构安全监测方法、***、设备及介质
CN115951361A (zh) 基于大数据平台的海上风电桩的智能地形扫测方法
Wang et al. The replacement of dysfunctional sensors based on the digital twin method during the cutter suction dredger construction process
JP3562494B2 (ja) ソーナー探知範囲予察可視化システム、方法及びプログラム
CN115540828A (zh) 一种基于壁面传感器的内波预报方法
US20200293704A1 (en) Method, a system and a computer program product for monitoring remote infrastructure networks
CN108594241B (zh) 一种auv声隐身态势评估方法
Spiliopoulos et al. A Big Data framework for Modelling and Simulating high-resolution hydrodynamic models in sea harbours
Xing et al. Sensor placement for robust burst identification in water systems: Balancing modeling accuracy, parsimony, and uncertainties
Teng et al. Review of intelligent detection and health assessment of underwater structures
CN107220425A (zh) 海缆作业操控方法、装置及服务终端
CN116579214A (zh) 一种基于数字孪生的三维可视化桥墩监测***及方法
KR102561270B1 (ko) 구조 구난을 위한 해상 표류물의 위치 추정 방법 및 장치
KR20230143444A (ko) 해상 표류체의 표류경로 추정시스템 및 방법
CN114739891A (zh) 一种漂浮式海上风电系泊***腐蚀状态检测评估***
Kimpton et al. A rapid simplified method for determining tsunami inundation extent based on energy conservation
Liu et al. Trajectory risk cognition of ship collision accident based on fusion of multi-model spatial data
CN114722130B (zh) 一种基于gis***的数字海洋海底电缆状态监控方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21932511

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21932511

Country of ref document: EP

Kind code of ref document: A1