Early warning method, device and equipment based on multi-source pipeline data fusion and storage medium

Abstract

The application discloses a kind of early warning method, device, equipment and storage medium based on multi-source pipeline data fusion.The method comprises: obtaining multiple multi-source pipeline data, wherein the multi-source pipeline data includes at least two of sensor data, pipeline state data, working condition data and environmental data;The multi-source pipeline data is input into the target prediction model, and the accident prediction data corresponding to the target fusion data is determined based on the model output result;Wherein, the target accident prediction model is a support vector machine model obtained by pre-training based on sample fusion data;Determine the target early warning strategy corresponding to the accident prediction data, and carry out safety warning based on the target early warning strategy.Improve the rapidity and correctness of pipeline safety warning.

Description

Technical Field

[0001] This invention relates to the field of pipeline safety detection technology, and in particular to an early warning method, device, equipment and storage medium based on multi-source pipeline data fusion. Background Technology

[0002] Long-distance oil and gas pipelines have played a vital role in the country's regional energy supply. To ensure the safety of pipeline transportation, various safety monitoring technologies have been developed and implemented, including fiber optic early warning, geological disaster monitoring, and video surveillance.

[0003] Because each early warning and monitoring system is relatively independent, with inconsistent data architecture, communication standards, and early warning level definitions, safety early warnings can only be provided through single pipeline data, resulting in low accuracy of pipeline safety early warnings. Summary of the Invention

[0004] This invention provides an early warning method, device, equipment, and storage medium based on multi-source pipeline data fusion to solve the problem of low accuracy in pipeline safety early warning.

[0005] According to one aspect of the present invention, an early warning method based on multi-source pipeline data fusion is provided, the method comprising:

[0006] Acquire multi-source pipeline data, wherein the multi-source pipeline data includes at least two of the following: sensor data, pipeline status data, operating condition data, and environmental data;

[0007] The multi-source pipeline data is input into the target prediction model, and the accident prediction data corresponding to the multi-source pipeline data is determined based on the model output; wherein, the target accident prediction model is a support vector machine model pre-trained based on sample fusion data;

[0008] Determine the target early warning strategy corresponding to the accident prediction data, and conduct safety early warning based on the target early warning strategy.

[0009] According to another aspect of the present invention, an early warning device based on multi-source pipeline data fusion is provided, the device comprising:

[0010] The data acquisition module is used to acquire various multi-source pipeline data, wherein the multi-source pipeline data includes at least two of the following: sensor data, pipeline status data, operating condition data, and environmental data.

[0011] The data prediction module is used to input the multi-source pipeline data into the target prediction model and determine the accident prediction data corresponding to the multi-source pipeline data based on the model output results; wherein, the target accident prediction model is a support vector machine model pre-trained based on sample fusion data;

[0012] The safety early warning module is used to determine the target early warning strategy corresponding to the accident prediction data, and to conduct safety early warning based on the target early warning strategy.

[0013] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0014] At least one processor; and

[0015] A memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the early warning method based on multi-source pipeline data fusion as described in any embodiment of the present invention.

[0017] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions, the computer instructions being configured to cause a processor to execute and implement the early warning method based on multi-source pipeline data fusion as described in any embodiment of the present invention.

[0018] The technical solution of this invention acquires multi-source pipeline data, including at least two of sensor data, pipeline status data, operating condition data, and environmental data. This acquisition provides more comprehensive information. The multi-source pipeline data is then input into a target prediction model. Based on the model's output, accident prediction data corresponding to the multi-source pipeline data is determined. The target accident prediction model is a support vector machine model pre-trained based on sample fusion data. This accurately determines the pipeline accident prediction data. Finally, a target early warning strategy corresponding to the accident prediction data is determined, and a safety early warning is issued based on this strategy. This solves the problem of low accuracy in pipeline safety early warning and improves the speed and accuracy of pipeline safety early warning.

[0019] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart of an early warning method based on multi-source pipeline data fusion according to Embodiment 1 of the present invention;

[0022] Figure 2 This is a flowchart of an early warning method based on multi-source pipeline data fusion provided in Embodiment 2 of the present invention;

[0023] Figure 3 This is a schematic diagram of the structure of an early warning device based on multi-source pipeline data fusion according to Embodiment 3 of the present invention;

[0024] Figure 4 This is a schematic diagram of the structure of an electronic device that implements the early warning method based on multi-source pipeline data fusion according to an embodiment of the present invention. Detailed Implementation

[0025] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0026] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0027] Example 1

[0028] Figure 1 This is a flowchart illustrating an early warning method based on multi-source pipeline data fusion, as provided in Embodiment 1 of the present invention. This embodiment is applicable to pipeline-based early warning situations. The method can be executed by an early warning device based on multi-source pipeline data fusion, which can be implemented in hardware and / or software and can be configured in an electronic device. Figure 1 As shown, the method includes:

[0029] S110. Acquire multi-source pipeline data, wherein the multi-source pipeline data includes at least two of the following: sensor data, pipeline status data, operating condition data, and environmental data.

[0030] Multi-source pipeline data can be understood as pipeline data obtained through multiple acquisition sources.

[0031] Specifically, data is collected through various sensors installed on the pipeline system (such as temperature sensors, pressure sensors, flow sensors, etc.). Regular inspections, maintenance records, historical data analysis, or advanced detection technologies (such as ultrasonic testing and radiographic testing) are used to obtain data describing the pipeline's condition, such as pipe material, thickness, corrosion level, and leakage history. Data related to the pipeline system's operating conditions, such as fluid type, flow rate, pressure, and temperature, are also obtained through regular inspections, maintenance records, historical data analysis, or advanced detection technologies (such as ultrasonic testing and radiographic testing). Data on external environmental factors affecting the pipeline system, such as geological conditions, climate conditions, and human activities, are obtained through meteorological stations, geological surveys, or remote sensing technology.

[0032] Optionally, the method further includes: preprocessing the multi-source pipeline data, and updating the multi-source pipeline data based on the preprocessed data, wherein the preprocessing includes at least one of data time alignment processing, data spatial alignment, data transformation processing, unified encoding processing, data denoising processing, and discretization processing.

[0033] Specifically, for pipeline data with different time resolutions or time offsets, time interpolation, time synchronization algorithms, or manual timestamp adjustment are used to align the pipeline data. Geographic Information System (GIS) technology can be used to convert the spatial coordinates of different data sources into a unified coordinate system, or spatial interpolation methods can be used to fill spatial data gaps to achieve spatial alignment. Pipeline data in different formats should be converted to a unified format. Using coding standards or custom coding schemes, a unique code should be assigned to each category, attribute, or entity in the multi-source pipeline data. Unified coding makes the data easier to manage and analyze, improving data processing efficiency. Statistical methods (such as median filtering and mean filtering) or machine learning algorithms (such as anomaly detection models) should be used to identify and remove noise. Multi-source pipeline data should be divided into different intervals or categories, and a discrete label should be assigned to each interval or category.

[0034] S120. Input the multi-source pipeline data into the target prediction model, and determine the accident prediction data corresponding to the multi-source pipeline data based on the model output results; wherein, the target accident prediction model is a support vector machine model pre-trained based on sample fusion data.

[0035] Among them, accident prediction data can be understood as safety accident data predicted by the model.

[0036] Specifically, the multi-source pipeline data is input into the target prediction model, which outputs a prediction result indicating the likelihood of an accident occurring given the data, such as a probability value or classification label. Based on the model's output, the accident prediction data corresponding to the multi-source pipeline data is determined.

[0037] S130. Determine the target early warning strategy corresponding to the accident prediction data, and conduct safety early warning based on the target early warning strategy.

[0038] Among them, the target early warning strategy can be understood as a systematic plan developed to prevent, identify, assess and respond to potential risks or threats.

[0039] Specifically, in-depth analysis of accident prediction data is conducted to determine the likelihood, type, scale, and potential scope of impact of accidents. Key risk factors and potential hazards are identified. Based on the analysis results of the accident prediction data, corresponding target-oriented early warning strategies are developed.

[0040] Optionally, determining the target early warning strategy corresponding to the accident prediction data includes:

[0041] Determine the accident type and / or accident level corresponding to the accident prediction data;

[0042] Based on the accident type and / or the accident level, a target early warning strategy corresponding to the accident prediction data is determined.

[0043] Among these, "accident type" can be understood as the type of safety accident, and "accident level" can be understood as the severity level of the accident.

[0044] Specifically, based on the results of data analysis, possible accident types are identified, such as leaks, fires, and collapses. Accidents are then classified into different levels based on their potential impact, severity, and urgency, such as minor, moderate, severe, and extremely severe. Assessing the accident level helps determine the priority and urgency of early warning strategies. Specific targeted early warning strategies are then developed for each identified accident type and level. These strategies may include real-time monitoring, data analysis, risk assessment, information dissemination, and emergency preparedness.

[0045] Optionally, the target early warning strategy includes a first early warning strategy corresponding to the accident type and / or a second early warning strategy corresponding to the accident level; determining the target early warning strategy corresponding to the accident prediction data based on the accident type and / or the accident level includes:

[0046] A target early warning strategy corresponding to the accident prediction data is determined based on the first early warning strategy and / or the second early warning strategy.

[0047] The first early warning strategy consists of early warning measures and contingency plans developed for specific accident types. The second early warning strategy consists of early warning measures and contingency plans developed based on the accident level.

[0048] Specifically, either the first early warning strategy or the second early warning strategy can be used as the target early warning strategy. Alternatively, the first and second early warning strategies can be combined to obtain a target early warning strategy corresponding to the accident prediction data.

[0049] For example, leak detection strategies may include monitoring the sealing of critical equipment and pipelines to promptly detect leaks. Installing leak detection alarm systems will trigger an alarm immediately upon detection. Developing leak emergency response plans will clearly define emergency procedures and responsible personnel. Fire detection strategies may include installing fire alarm systems to monitor fire hazards in real time. Regularly inspecting fire protection facilities to ensure their proper functioning. Developing fire emergency plans, including evacuation routes and firefighting methods. Accident severity levels are typically classified based on factors such as the severity and scope of the accident.

[0050] For example, a minor accident early warning strategy may include taking routine inspections and monitoring measures to promptly identify and address potential safety hazards. Minor accidents should be recorded and analyzed to summarize lessons learned and prevent similar accidents from recurring. A general accident early warning strategy may include strengthening safety inspections and risk assessments to promptly identify and eliminate general safety hazards. Emergency response plans for general accidents should be developed, clearly defining emergency response procedures and responsible personnel. Emergency drills should be organized to improve employees' ability to respond to general accidents. A serious accident early warning strategy may include establishing a dedicated safety management organization responsible for the early warning and emergency management of serious accidents. Safety monitoring and early warning systems should be strengthened to improve the accuracy and timeliness of early warnings. Detailed and comprehensive emergency response plans for serious accidents should be developed, including emergency response, rescue, and recovery measures. A particularly serious accident early warning strategy may include further strengthening safety management measures based on serious accident early warnings. A linkage mechanism with relevant departments and agencies should be established to jointly respond to particularly serious accidents.

[0051] Regularly organize cross-departmental and cross-regional emergency drills to improve the ability to respond to particularly serious accidents.

[0052] Optionally, preset early warning strategies corresponding to different accident types and preset early warning strategies corresponding to different accident levels can be set in advance.

[0053] The technical solution of this invention acquires multi-source pipeline data, including at least two of sensor data, pipeline status data, operating condition data, and environmental data. This acquisition provides more comprehensive information. The multi-source pipeline data is then input into a target prediction model. Based on the model's output, accident prediction data corresponding to the multi-source pipeline data is determined. The target accident prediction model is a support vector machine model pre-trained based on sample fusion data. This accurately determines the pipeline accident prediction data. Finally, a target early warning strategy corresponding to the accident prediction data is determined, and a safety early warning is issued based on this strategy. This solves the problem of low accuracy in pipeline safety early warning and improves the speed and accuracy of pipeline safety early warning.

[0054] Example 2

[0055] Figure 2 This is a flowchart of an early warning method based on multi-source pipeline data fusion provided in Embodiment 2 of the present invention. This embodiment is a further optimization of the input of the multi-source pipeline data into the target prediction model in the above embodiment. Optionally, before inputting the multi-source pipeline data into the target prediction model, the method further includes: acquiring a preset number of sample pipeline association data; constructing a sample fusion dataset based on the sample pipeline association data; dividing the sample fusion dataset into a training set and a test set; wherein the sample pipeline association data includes sample pipeline data and sample accident labels corresponding to the sample pipeline data; iteratively training a pre-established support vector machine model using the training set; determining the model loss based on the model output label obtained in each training iteration and the sample accident label; adjusting the model parameters based on the model loss; testing the model after each training iteration using the test set to obtain model prediction performance indicators; and determining the target prediction model based on multiple model prediction performance indicators.

[0056] like Figure 2 As shown, the method includes:

[0057] S210. Acquire multi-source pipeline data, wherein the multi-source pipeline data includes at least two of the following: sensor data, pipeline status data, operating condition data, and environmental data.

[0058] S220. Obtain a preset number of sample pipeline association data, construct a sample fusion dataset based on the sample pipeline association data, and divide the sample fusion dataset into a training set and a test set.

[0059] The sample pipeline association data includes sample pipeline data and sample accident tags corresponding to the sample pipeline data.

[0060] Specifically, a preset number of sample pipeline associated data are fused to construct a sample fusion dataset.

[0061] Optionally, constructing the sample fusion dataset based on the sample pipeline association data includes:

[0062] The pre-set number of sample pipeline correlation data are integrated and processed to obtain initial fused data;

[0063] Extract key fusion data from the initial fusion data, and construct a sample fusion dataset based on the key fusion data.

[0064] The initial fused data can be understood as the data set after preliminary integration and processing. The key fused data can be understood as the data set after eliminating redundant information.

[0065] Specifically, a predetermined number of sample pipeline correlation data are initially integrated to obtain initial fused data. Redundant and unimportant features in the initial fused data are removed, and the set of features that contribute most to the pipeline safety status, i.e., key fused data, is retained. A sample fused dataset is then constructed based on the dataset of key fused data.

[0066] Optionally, extracting key fusion data from the initial fusion data includes:

[0067] The key fusion data is extracted from the initial fusion data based on the attribute reduction algorithm.

[0068] Among them, attribute reduction algorithms include at least one of dependency-based reduction, discrimination matrix-based reduction, and heuristic algorithms.

[0069] Specifically, the key fusion data is extracted from the initial fusion data using attribute reduction algorithms based on rough set theory. The initial fusion data is then input into the selected attribute reduction algorithm. The algorithm analyzes the relationship between each feature and the target variable and evaluates the importance of each feature. Based on the algorithm's output, the most important features are selected to form the key fusion dataset. Cross-validation, hold-out, or other validation strategies are used to evaluate the effectiveness of the key fusion dataset. The performance of the model trained using the key fusion dataset is compared with that trained on the original dataset to ensure that the key fusion dataset does not lose too much important information.

[0070] S230. The pre-established support vector machine model is iteratively trained using the training set. The model loss is determined based on the model output label and the sample accident label obtained in each training session. The model parameters are adjusted based on the model loss.

[0071] Specifically, select a suitable kernel function (e.g., linear kernel, radial basis function kernel, etc.) and set initial parameters. Input the samples in the training set one by one or in batches into the support vector machine model. Calculate the predicted label for each sample based on the current model parameters. Compare the predicted label with the actual accident label of the sample and calculate the model loss. Based on the model loss, adjust the model parameters using an optimization algorithm (e.g., gradient descent). Repeat the above steps until a predetermined number of iterations is reached or the model loss converges below a certain threshold.

[0072] S240. The model is tested each time it has been trained using the test set to obtain model prediction performance metrics, and the target prediction model is determined based on multiple model prediction performance metrics.

[0073] Specifically, after each training iteration, the model is tested using a test set. Samples from the test set are input one by one into the trained support vector machine model to obtain the model's predicted label for each sample. Based on the model's predicted label and the sample's predicted label, a series of prediction performance metrics are calculated, such as accuracy, recall, and F1 score. The prediction performance metrics of the model obtained after each training iteration are recorded. Based on a comprehensive consideration of multiple prediction performance metrics, the model with the best performance is selected as the target prediction model.

[0074] S250. Input the multi-source pipeline data into the target prediction model, and determine the accident prediction data corresponding to the multi-source pipeline data based on the model output results.

[0075] The target accident prediction model is a support vector machine model pre-trained based on sample fusion data.

[0076] S260. Determine the target early warning strategy corresponding to the accident prediction data, and conduct safety early warning based on the target early warning strategy.

[0077] The technical solution of this invention involves acquiring a preset number of sample pipeline association data, constructing a sample fusion dataset based on the sample pipeline association data, and dividing the sample fusion dataset into a training set and a test set. The sample pipeline association data includes sample pipeline data and sample accident labels corresponding to the sample pipeline data. A pre-established support vector machine model is iteratively trained using the training set. The model loss is determined based on the model output label obtained in each training iteration and the sample accident label, and the model parameters are adjusted based on the model loss. The model is then tested using the test set to obtain model prediction performance metrics. Based on multiple model prediction performance metrics, a target prediction model is determined. This target prediction model can accurately predict pipeline accidents, allowing for proactive preventative or remedial measures, thereby reducing the probability and impact of accidents.

[0078] As an optional example of Embodiment 1 of the present invention, the early warning method for multi-source pipeline data fusion in this embodiment specifically includes the following steps:

[0079] The process includes steps such as data source diversification and integration, data preprocessing, data feature extraction and selection, and early warning event identification and analysis.

[0080] Specifically, the pipeline early warning platform collects multi-source pipeline data from existing pipeline safety early warning data sources, and unifies data standards and interface technical specifications. It uses support vector machines to eliminate noise in early warning data, and achieves multi-source data fusion through time alignment, spatial alignment, data transformation, and unified coding. Utilizing a rough set-support vector machine salient type model, it integrates support vector machines and rough sets, leveraging the advantages of rough set theory in handling large amounts of data and eliminating redundant information to extract the set of feature indicators that make the greatest contribution to pipeline safety. It extracts the most representative and discriminative features from a large amount of data, analyzes, infers, and discovers relationships between data, solving the problem of selecting feature indicators (sensitive indicators) in pipeline safety salient prediction research, and providing valuable feature information and data support. A comprehensive evaluation method based on relevant data fusion and complementary data fusion is adopted to coordinate, optimize, and comprehensively process multi-source information. This intelligently processes multi-source information from multiple measurement devices, mimicking the comprehensive information processing capabilities of experts, and trains and optimizes it based on historical data and known accident cases to improve the accuracy and timeliness of early warnings.

[0081] Furthermore, the data interface can be divided into basic information of monitoring and sensing devices, status monitoring information and early warning information, including but not limited to sensors, monitoring equipment, manual inspections, etc., and these data can cover pipeline status information, operating data, environmental parameters, etc.

[0082] Furthermore, the noise reduction method based on support vector machine-based early warning data includes steps such as data cleaning, data calibration, outlier detection, and missing value imputation to eliminate the influence of ordinary noise, abnormal data, and missing data in the monitoring data, ensuring the accuracy and integrity of the data; and introduces discretization processing of continuous conditional attributes to convert continuous numerical data into discrete categories that are more suitable for subsequent analysis.

[0083] For example, the temporal and spatial alignment methods in the data fusion step can be one or more of recurrent neural networks, convolutional neural networks, graph neural networks, and attention mechanisms; the data transformation and unified encoding methods can be one or more of linear and nonlinear support vector machines and regression, decision trees, hidden Markov models, random forests, neural networks, deep learning networks, boosting, feature selection, clustering, approximation, and dynamic programming.

[0084] Furthermore, rough set theory is used for attribute reduction and rule extraction to identify and remove redundant or irrelevant features from the preprocessed data, extract key fusion data from the initial fusion data, and construct a sample fusion dataset based on the key fusion data.

[0085] Furthermore, the simplified sample fusion dataset is divided into training samples and test samples. The training samples are used to build a support vector machine model, and the model parameters are optimized through the training process to capture complex patterns and relationships in the data.

[0086] Furthermore, the comprehensive evaluation method based on relevant data fusion and complementary data fusion effectively fuses relevant and complementary data provided by similar / dissimilar sensors in the pipeline safety monitoring system. This integrates multi-source information from various aspects of the pipeline environment into a single framework. Historical data and known accident cases can be used to train and test the support vector machine model, and comprehensive analysis yields reliable feature vectors of pipeline safety status. This provides a more accurate, comprehensive, and complete description of the object and environment, achieving a comprehensive evaluation of pipeline safety status.

[0087] The technical solution of this invention can fuse and analyze multi-source early warning data from various pipeline safety early warning systems. By preprocessing the monitoring data to ensure its accuracy and completeness, it extracts the data's information characteristics and variation patterns. A multivariate data fusion structure model is applied to fuse multi-sensor monitoring data and complementary data to analyze the pipeline safety status, enabling users to quickly understand and take appropriate countermeasures. Through multi-source data fusion early warning analysis, early warning information from different sources, modes, media, times, spaces, and representations can be organically combined. This eliminates redundancy and contradictions between different early warning information, complements them, and reduces uncertainty, forming a more accurate and relatively complete description of the pipeline safety status. It obtains precise features that are impossible to obtain with a single early warning technology, thereby improving the speed and accuracy of intelligent pipeline safety early warning analysis and reducing decision-making risks.

[0088] Example 3

[0089] Figure 3 This is a schematic diagram of an early warning device based on multi-source pipeline data fusion, provided in Embodiment 3 of the present invention. Figure 3 As shown, the device includes: a data acquisition module 310, a data prediction module 320, and a safety early warning module 330.

[0090] The data acquisition module 310 is used to acquire multiple sources of pipeline data, including at least two of sensor data, pipeline status data, operating condition data, and environmental data. The data prediction module 320 is used to input the multi-source pipeline data into a target prediction model and determine the accident prediction data corresponding to the multi-source pipeline data based on the model output. The target accident prediction model is a support vector machine model pre-trained based on sample fusion data. The safety early warning module 330 is used to determine the target early warning strategy corresponding to the accident prediction data and conduct safety early warning based on the target early warning strategy.

[0091] The technical solution of this invention acquires multi-source pipeline data, including at least two of sensor data, pipeline status data, operating condition data, and environmental data. This acquisition provides more comprehensive information. The multi-source pipeline data is then input into a target prediction model. Based on the model's output, accident prediction data corresponding to the multi-source pipeline data is determined. The target accident prediction model is a support vector machine model pre-trained based on sample fusion data. This accurately determines the pipeline accident prediction data. Finally, a target early warning strategy corresponding to the accident prediction data is determined, and a safety early warning is issued based on this strategy. This solves the problem of low accuracy in pipeline safety early warning and improves the speed and accuracy of pipeline safety early warning.

[0092] Optionally, the device further includes:

[0093] The dataset construction module is used to acquire a preset number of sample pipeline association data before inputting the multi-source pipeline data into the target prediction model, construct a sample fusion dataset based on the sample pipeline association data, and divide the sample fusion dataset into a training set and a test set; wherein, the sample pipeline association data includes sample pipeline data and sample accident labels corresponding to the sample pipeline data;

[0094] The model training module is used to iteratively train the pre-established support vector machine model using the training set, determine the model loss based on the model output label and the sample accident label obtained in each training, and adjust the model parameters based on the model loss.

[0095] The model determination module is used to test the model after each training session using the test set to obtain model prediction performance metrics, and to determine the target prediction model based on multiple model prediction performance metrics.

[0096] Optionally, the dataset construction module includes:

[0097] The data fusion unit is used to integrate and process a preset number of sample pipeline correlation data to obtain initial fused data;

[0098] The dataset construction unit is used to extract key fusion data from the initial fusion data and construct a sample fusion dataset based on the key fusion data.

[0099] Optionally, the dataset construction unit is specifically used for:

[0100] Key fusion data are extracted from the initial fusion data based on the attribute reduction algorithm.

[0101] Optionally, the security warning module includes:

[0102] An accident determination unit is used to determine the accident type and / or accident level corresponding to the accident prediction data;

[0103] The early warning strategy determination unit is used to determine a target early warning strategy corresponding to the accident prediction data based on the accident type and / or the accident level.

[0104] Optionally, the target early warning strategy includes a first early warning strategy corresponding to the accident type and / or a second early warning strategy corresponding to the accident level; correspondingly, the early warning strategy determination unit is specifically used for:

[0105] A target early warning strategy corresponding to the accident prediction data is determined based on the first early warning strategy and / or the second early warning strategy.

[0106] The device further includes:

[0107] A preprocessing module is used to preprocess the multi-source pipeline data and update the multi-source pipeline data based on the preprocessed data. The preprocessing includes at least one of data time alignment processing, data spatial alignment, data transformation processing, unified encoding processing, data denoising processing, and discretization processing.

[0108] The early warning device based on multi-source pipeline data fusion provided in the embodiments of the present invention can execute the early warning method based on multi-source pipeline data fusion provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0109] Example 4

[0110] Figure 4A schematic diagram of an electronic device 10 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0111] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0112] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0113] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as early warning based on multi-source pipeline data fusion.

[0114] In some embodiments, the method for early warning based on multi-source pipeline data fusion can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the method for early warning based on multi-source pipeline data fusion described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the method for early warning based on multi-source pipeline data fusion by any other suitable means (e.g., by means of firmware).

[0115] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0116] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0117] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0118] To provide interaction with a service recipient, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the service recipient; and a keyboard and pointing device (e.g., a mouse or trackball) through which the service recipient can provide input to the electronic device. Other types of devices can also be used to provide interaction with the service recipient; for example, feedback provided to the service recipient can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the service recipient can be received in any form (including voice input, speech input, or tactile input).

[0119] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., service-seeking computers with a graphical service-seeking interface or a web browser through which the service-seeking party can interact with the implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0120] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0121] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0122] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. An early warning method based on multi-source pipeline data fusion, characterized in that, include: Acquire multi-source pipeline data, wherein the multi-source pipeline data includes at least two of the following: sensor data, pipeline status data, operating condition data, and environmental data; The multi-source pipeline data is input into the target prediction model, and the accident prediction data corresponding to the multi-source pipeline data is determined based on the model output; wherein, the target accident prediction model is a support vector machine model pre-trained based on sample fusion data; Determine the target early warning strategy corresponding to the accident prediction data, and conduct safety early warning based on the target early warning strategy.

2. The method according to claim 1, characterized in that, Before inputting the multi-source pipeline data into the target prediction model, the following steps are also included: A preset number of sample pipeline association data are obtained, and a sample fusion dataset is constructed based on the sample pipeline association data. The sample fusion dataset is divided into a training set and a test set. The sample pipeline association data includes sample pipeline data and sample incident labels corresponding to the sample pipeline data. The pre-established support vector machine model is iteratively trained using the training set. The model loss is determined based on the model output label and the sample accident label obtained in each training session. The model parameters are then adjusted based on the model loss. The model is tested on the test set for each completed training session to obtain model prediction performance metrics, and the target prediction model is determined based on multiple model prediction performance metrics.

3. The method according to claim 2, characterized in that, The construction of the sample fusion dataset based on the sample pipeline association data includes: The pre-set number of sample pipeline correlation data are integrated and processed to obtain initial fused data; Extract key fusion data from the initial fusion data, and construct a sample fusion dataset based on the key fusion data.

4. The method according to claim 3, characterized in that, The extraction of key fusion data from the initial fusion data includes: Key fusion data are extracted from the initial fusion data based on the attribute reduction algorithm.

5. The method according to claim 1, characterized in that, The determination of the target early warning strategy corresponding to the accident prediction data includes: Determine the accident type and / or accident level corresponding to the accident prediction data; Based on the accident type and / or the accident level, a target early warning strategy corresponding to the accident prediction data is determined.

6. The method according to claim 5, characterized in that, The target early warning strategy includes a first early warning strategy corresponding to the accident type and / or a second early warning strategy corresponding to the accident level; Determining a target early warning strategy corresponding to the accident prediction data based on the accident type and / or the accident level includes: A target early warning strategy corresponding to the accident prediction data is determined based on the first early warning strategy and / or the second early warning strategy.

7. The method according to claim 1, characterized in that, Also includes: The multi-source pipeline data is preprocessed, and the multi-source pipeline data is updated based on the preprocessed data. The preprocessing includes at least one of the following: data time alignment, data spatial alignment, data transformation, unified encoding, data denoising, and discretization.

8. An early warning device based on multi-source pipeline data fusion, characterized in that, include: The data acquisition module is used to acquire various multi-source pipeline data, wherein the multi-source pipeline data includes at least two of the following: sensor data, pipeline status data, operating condition data, and environmental data. The data prediction module is used to input the multi-source pipeline data into the target prediction model and determine the accident prediction data corresponding to the multi-source pipeline data based on the model output results; wherein, the target accident prediction model is a support vector machine model pre-trained based on sample fusion data; The safety early warning module is used to determine the target early warning strategy corresponding to the accident prediction data, and to conduct safety early warning based on the target early warning strategy.

9. An electronic device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the early warning method based on multi-source pipeline data fusion as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the early warning method based on multi-source pipeline data fusion as described in any one of claims 1-7.

Images