Risk prevention and control platform for underground utility tunnel

Abstract

This invention discloses a risk prevention and control platform for underground utility tunnels, relating to the field of Internet of Things (IoT) technology. The platform includes: a fixed-point monitoring module for acquiring multi-source monitoring datasets using IoT; a risk assessment module for acquiring multiple risk coefficients; a comprehensive risk calculation module for obtaining an overall risk coefficient; and a risk early warning module for providing early warnings of tunnel operation risks. This invention solves the technical problems in existing technologies, such as the difficulty in obtaining accurate multi-source monitoring data for underground utility tunnels, the lack of effective methods for accurately assessing the operational risks of each pipeline using monitoring data, and the inability to provide timely and accurate early warnings based on risk assessment results. It achieves the technical effects of accurately acquiring multi-source monitoring data for underground utility tunnels, accurately assessing the operational risks of each pipeline and calculating the overall risk coefficient, realizing timely and accurate risk early warnings, and effectively ensuring the safe and stable operation of underground utility tunnels.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet of Things, in particular to the technical field of a risk prevention and control platform for underground comprehensive pipe galleries. BACKGROUND

[0002] In the field of risk prevention and control of underground comprehensive pipe galleries, the existing technology mainly relies on manual inspection and simple monitoring equipment. Manual inspection has problems such as low efficiency and untimely detection, and it is difficult to monitor multiple public pipelines in the pipe gallery in real time. Simple monitoring equipment can only obtain a small amount of data and cannot fully reflect the pipeline operation status. These traditional methods have many limitations when faced with the complex environment and diverse pipelines in the pipe gallery. Due to the closed space and complex environment of underground comprehensive pipe galleries, and the inclusion of power, communication, water supply and drainage, and other pipelines, traditional methods cannot accurately obtain the operation data of each pipeline, making it difficult to accurately assess and timely warn of the operation risk of the pipe gallery, and cannot meet the needs of ensuring the safe and stable operation of the pipe gallery. SUMMARY

[0003] The present application provides a risk prevention and control platform for underground comprehensive pipe galleries, which solves the technical problems in the prior art that it is difficult to obtain accurate multi-source monitoring data for underground comprehensive pipe galleries, there is a lack of effective methods to accurately assess the operation risk of each pipeline using monitoring data, and it is not possible to timely and accurately warn according to the risk assessment results.

[0004] In view of the above problems, the present application provides a risk prevention and control platform for underground comprehensive pipe galleries.

[0005] The present application provides a risk prevention and control platform for underground comprehensive pipe galleries, which comprises:

[0006] A fixed-point monitoring module for using Internet of Things to perform fixed-point monitoring on multiple public pipelines in the underground comprehensive pipe gallery and obtain a multi-source monitoring data set; an operation risk assessment module for performing operation risk assessment based on the multi-source monitoring data set and obtaining multiple risk coefficients; a comprehensive risk calculation module for performing comprehensive risk calculation based on the multiple risk coefficients and obtaining an overall risk coefficient; and an operation risk warning module for performing pipe gallery operation risk warning if the overall risk coefficient is greater than a predetermined risk coefficient threshold.

[0007] One or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0008] The point monitoring module is used for monitoring multiple public pipelines in the underground comprehensive pipe gallery by using the Internet of Things to obtain a multi-source monitoring data set; the operation risk assessment module is used for respectively performing operation risk assessment according to the multi-source monitoring data set to obtain multiple risk coefficients; the comprehensive risk calculation module is used for performing comprehensive risk calculation based on the multiple risk coefficients to obtain an overall risk coefficient; and the operation risk early warning module is used for performing pipe gallery operation risk early warning if the overall risk coefficient is greater than a predetermined risk coefficient threshold. The technical effects of accurately obtaining multi-source monitoring data of the underground comprehensive pipe gallery, accurately assessing operation risks of each pipeline and calculating an overall risk coefficient, timely and accurately performing risk early warning and effectively guaranteeing safe and stable operation of the underground comprehensive pipe gallery are achieved. BRIEF DESCRIPTION OF DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0010] Figure 1 The structural schematic diagram of the risk prevention and control platform for the underground comprehensive pipe gallery provided by the embodiments of the present application is shown.

[0011] Figure 2 The structural schematic diagram of the plug-in evaluation unit in the risk prevention and control platform for the underground comprehensive pipe gallery provided by the embodiments of the present application is shown.

[0012] Explanation of reference signs: point monitoring module 10, operation risk assessment module 20, comprehensive risk calculation module 30, operation risk early warning module 40. DETAILED DESCRIPTION

[0013] The present application provides a risk prevention and control platform for the underground comprehensive pipe gallery to solve the technical problems in the prior art that it is difficult to obtain accurate multi-source monitoring data of the underground comprehensive pipe gallery, there is a lack of effective method to accurately assess operation risks of each pipeline by using monitoring data, and it is impossible to timely and accurately perform early warning according to the risk assessment results.

[0014] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings of the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0015] Embodiments, such as Figure 1As shown, the present application provides a risk prevention and control platform for underground utility tunnels, which comprises:

[0016] A fixed-point monitoring module 10 is configured to utilize the Internet of Things to perform fixed-point monitoring on a plurality of public pipelines in the underground utility tunnel and obtain a multi-source monitoring data set.

[0017] In the embodiments of the present application, the underground utility tunnel is a public tunnel built underground in a city and used for centrally laying two or more municipal pipelines such as power, communication, broadcast television, water supply, drainage, heat supply, and gas pipelines. The multi-source monitoring data set is a data set obtained by utilizing the Internet of Things to perform fixed-point monitoring on a plurality of public pipelines in the underground utility tunnel.

[0018] Specifically, first, in the underground utility tunnel, various sensors are reasonably deployed on a plurality of public pipelines such as power, communication, water supply, drainage, gas, and heat supply pipelines. These sensors have different monitoring functions, such as monitoring pressure, temperature, flow, displacement, etc. Then, the sensors start working and collect relevant data of the pipelines at their locations in real time. Subsequently, through the Internet of Things communication technology, the data collected by the sensors dispersed in the tunnel are aggregated and encoded and packaged according to a certain protocol, such as the MQTT (a lightweight message transmission protocol based on the publish / subscribe mode) protocol, which has the characteristics of lightweight, low power consumption, and is suitable for transmitting data in unstable network environments, and is very suitable for complex environments such as underground utility tunnels. Finally, the data are transmitted to the data processing center, where they are unpacked and sorted, thereby obtaining a multi-source monitoring data set containing multiple types and multiple dimensions of information, which provides a comprehensive and accurate data basis for subsequent risk assessment of the public pipelines in the tunnel.

[0019] An operation risk assessment module 20 is configured to perform operation risk assessment on the multi-source monitoring data set respectively and obtain a plurality of risk coefficients.

[0020] In the embodiments of the present application, the plurality of risk coefficients are a series of numerical values obtained by performing operation risk assessment on a plurality of public pipelines in the underground utility tunnel respectively.

[0021] Specifically, first, a public pipeline is randomly selected from a plurality of public pipelines and set as a first public pipeline. Then, a first risk assessment plug-in is constructed for the first public pipeline. The first monitoring data set obtained in real time by the first public pipeline is input into the plug-in, which can output a first risk assessment coefficient added to the plurality of risk coefficients. The specific steps are described in detail in the pipeline selection unit, the plug-in assessment unit, and the risk coefficient addition unit.

[0022] Through the series of steps, the information in the multi-source monitoring data can be fully mined, the operation risk of each public pipeline can be accurately evaluated, and multiple reliable risk coefficients can be obtained, thereby providing strong support for subsequent comprehensive risk calculation and early warning.

[0023] The comprehensive risk calculation module 30 is configured to perform comprehensive risk calculation based on the multiple risk coefficients to obtain an overall risk coefficient.

[0024] In the embodiments of the present application, the comprehensive risk calculation is a process of calculating the multiple risk coefficients obtained by the operation risk evaluation module 20 according to the failure frequency and scale of each public pipeline in the preset time range based on the pipeline operation monitoring records. The overall risk coefficient is obtained by the comprehensive risk calculation and is used to reflect the numerical value of the overall operation risk of the underground comprehensive pipe gallery.

[0025] Specifically, first, the pipeline operation monitoring records in the preset time range are deeply mined. Then, based on the obtained multiple failure frequencies and multiple failure scale data, in-depth analysis is performed to determine multiple weight proportions. Finally, according to the determined multiple weight proportions, the comprehensive risk calculation is performed on the previously obtained multiple risk coefficients. The risk coefficient of each public pipeline is multiplied by the corresponding weight proportion, and then all the products are added to finally obtain the overall risk coefficient that can comprehensively reflect the overall operation risk of the underground comprehensive pipe gallery. The specific steps are described in detail in the monitoring record operation unit, the weight determination unit and the risk calculation unit.

[0026] Through the above steps, the comprehensive risk of the pipeline can be more accurately evaluated, thereby providing a reliable basis for subsequent risk early warning and prevention.

[0027] The operation risk early warning module 40 is configured to perform pipeline operation risk early warning if the overall risk coefficient is greater than a predetermined risk coefficient threshold.

[0028] In the embodiments of the present application, the predetermined risk coefficient threshold is a value determined according to multiple factors such as the pipeline operation safety standard, historical failure data, the importance of various pipelines and actual risk prevention and control requirements. The pipeline operation risk early warning is a series of operations performed by the operation risk early warning module 40 when the overall risk coefficient is greater than the predetermined risk coefficient threshold.

[0029] Specifically, first, the multiple public pipelines such as power, communication and water supply pipelines in the underground comprehensive pipe gallery are comprehensively and real-timely monitored by means of the Internet of Things technology to obtain a multi-source monitoring data set containing multiple information such as pressure, temperature and flow. These data are the basis for subsequent risk evaluation and can comprehensively reflect the operation state of various pipelines in the pipeline.

[0030] Then, the operation risk assessment is performed for each common pipeline. The process includes randomly selecting a common pipeline to construct a risk assessment plug-in, obtaining the monitoring index set of the pipeline, retrieving historical operation logs to obtain sample data and analyzing the sample risk coefficient set, training the feedforward neural network, and a series of operations to obtain the risk coefficient corresponding to each common pipeline. These risk coefficients reflect the potential operation risks of each pipeline.

[0031] Then, a comprehensive calculation is performed based on the plurality of risk coefficients obtained above. According to the operation monitoring records of the pipe gallery within a predetermined time range, the frequency and scale of failure of each common pipeline are counted, and the weight proportion corresponding to each risk coefficient is determined. Then, according to the corresponding calculation rules, the plurality of risk coefficients and the weight proportion are combined to finally obtain an overall risk coefficient that can reflect the overall operation risk status of the pipe gallery.

[0032] Finally, the overall risk coefficient calculated is compared with a predetermined risk coefficient threshold value set in advance. This threshold value is determined by technicians in the art based on safety standards, historical data, and actual needs of the pipe gallery operation, and other factors. If the overall risk coefficient is greater than the predetermined risk coefficient threshold value, the operation risk warning module 40 will immediately start the pipe gallery operation risk warning. The warning method can be to issue an audible and visual alarm, send a notification to the terminal device of the management personnel, etc., to ensure that relevant personnel can be aware of the risks existing in the pipe gallery in a timely manner, so as to take appropriate measures quickly and ensure the safe operation of the pipe gallery and its pipelines.

[0033] In one possible implementation, the fixed-point monitoring module 10 further includes:

[0034] The plurality of common pipelines at least includes power pipelines, communication pipelines, water supply pipelines, drainage pipelines, gas pipelines, and heating pipelines.

[0035] Specifically, the fixed-point monitoring module 10 uses Internet of Things technology to reasonably deploy sensors on various pipelines. For power pipelines, sensors for monitoring current, voltage, etc. are installed; communication pipelines are equipped with devices for monitoring signal strength and transmission rate; water supply pipelines are provided with water pressure and flow sensors; drainage pipelines focus on water level and flow; gas pipelines monitor gas concentration and pressure; heating pipelines monitor temperature and pressure, etc. These sensors continuously collect data to form a multi-source monitoring data set, allowing management personnel to understand the operation status of each pipeline in real time.

[0036] In one possible implementation, the operation risk assessment module 20 further includes:

[0037] A pipeline selection unit is configured to randomly select one common pipeline from the plurality of common pipelines as a first common pipeline.

[0038] a plug-in evaluation unit configured to construct a first risk evaluation plug-in based on the first common pipeline.

[0039] a risk coefficient adding unit configured to obtain a first monitoring data set of the first common pipeline, input the first risk evaluation plug-in, and output a first risk evaluation coefficient added to the plurality of risk coefficients.

[0040] In the embodiments of the present application, the first risk evaluation plug-in is obtained by obtaining a first monitoring index set of the first common pipeline, retrieving its historical operation log with the index set as a constraint, obtaining a plurality of sample monitoring data sets and analyzing a sample risk coefficient set, and training a feedforward neural network with these data until convergence.

[0041] Specifically, first, the pipeline selection unit randomly selects one of the power pipeline, the communication pipeline, the water supply pipeline, the drainage pipeline, the gas pipeline and the heating pipeline as the first common pipeline. The random selection method can ensure that each type of pipeline has an equal opportunity to be evaluated first, avoiding bias caused by fixed order selection, and covering all types of pipeline risk evaluation requirements comprehensively and evenly.

[0042] Next, the plug-in evaluation unit starts working. It first obtains a first monitoring index set of the first common pipeline. With these indexes as constraints, the historical operation log of the first common pipeline is retrieved to obtain a plurality of sample monitoring data sets. By analyzing the pipeline failure risk under different sample monitoring data, a sample risk coefficient set is obtained. Then, the sample monitoring data sets and the sample risk coefficient set are used to train the feedforward neural network until the network converges, thereby constructing the first risk evaluation plug-in. The specific steps are described in detail in the index set monitoring unit, the index set constraint unit and the evaluation plug-in obtaining unit.

[0043] Finally, the risk coefficient adding unit obtains the first monitoring data set of the first common pipeline in real time and inputs it into the constructed first risk evaluation plug-in. The plug-in processes the input data based on the learned rules and outputs the first risk evaluation coefficient. This coefficient represents the current risk level of the first common pipeline, and then the coefficient is added to the plurality of risk coefficient sets. Repeating the above process from pipeline selection to risk coefficient addition can evaluate the risk of all common pipelines in the pipe gallery.

[0044] By obtaining a plurality of risk coefficients, detailed and accurate data basis is provided for subsequent comprehensive evaluation of the overall risk of the pipe gallery.

[0045] In one possible implementation manner, as shown in Figure 2 The plug-in evaluation unit further includes:

[0046] An index set monitoring sub-unit is configured to obtain a first monitoring index set of the first public pipeline.

[0047] An index set constraint sub-unit is configured to take the first monitoring index set as a constraint, retrieve a historical operation log of the first public pipeline, obtain a plurality of sample monitoring data sets, and analyze pipeline failure risks under different sample monitoring data to obtain a sample risk coefficient set.

[0048] An evaluation plug-in obtaining sub-unit is configured to train a feedforward neural network to convergence by using the plurality of sample monitoring data sets and the sample risk coefficient set, and obtain the first risk evaluation plug-in.

[0049] In the embodiments of the present application, the first monitoring index set is a series of key parameter sets reflecting the operation state of the first public pipeline. The sample risk coefficient set is a series of risk coefficient sets obtained by analyzing pipeline failure risks under different sample monitoring data after obtaining the plurality of sample monitoring data sets.

[0050] Specifically, first, the index set monitoring sub-unit plays a role. Taking the power pipeline as an example, it will obtain current, voltage, temperature, resistance, etc. as the first monitoring index set. These indicators are key parameters reflecting the operation state of the power pipeline. Different indicators exhibit the working condition of the pipeline from different dimensions, providing a comprehensive data basis for subsequent risk assessment. For other types of public pipelines, such as communication pipelines, the index set may include signal strength, bit error rate, etc.; and the water supply pipeline may involve water pressure, water flow speed, etc.

[0051] Then, the index set constraint sub-unit starts working. It takes the obtained first monitoring index set as a constraint condition to retrieve the historical operation log of the first public pipeline. Through deep mining of historical data, a plurality of sample monitoring data sets can be obtained. These sample data cover the operation information of the pipeline under different working conditions, and the pipeline failure risks under different sample monitoring data are analyzed in detail. For example, the possibility and harm degree of the power pipeline failure under different conditions such as high current and unstable voltage are analyzed, so as to obtain the sample risk coefficient set. This step fully utilizes the rules in the historical data, and provides a strong basis for risk assessment.

[0052] Finally, the evaluation plug-in obtains sub-unit utilizes the obtained multiple sample monitoring data sets and sample risk coefficient sets to train the feedforward neural network. The multiple sample monitoring data sets and sample risk coefficient sets are input into the feedforward neural network as training data. The neurons inside the network transmit signals according to a specific connection mode. At the beginning of training, the weights and thresholds of the network are randomly initialized. During the training process, the feedforward neural network calculates according to the input sample monitoring data, outputs the predicted risk coefficient, and then compares the prediction result with the actual sample risk coefficient set to calculate the difference between the two through a loss function. Based on this difference, the backpropagation algorithm is used to adjust the weights and thresholds of the network to continuously optimize the fitting ability of the network to sample data. This process is repeated until the difference between the output result of the network and the actual sample risk coefficient reaches a minimum value, that is, the network converges, at which time the training of the feedforward neural network is completed, and the first risk evaluation plug-in for evaluating the running risk of the first public pipeline is obtained.

[0053] Through the plug-in, the running risk of the first public pipeline can be accurately evaluated according to the input real-time monitoring data, providing a precise evaluation tool for risk prevention and control of the underground comprehensive pipe gallery.

[0054] In a possible implementation manner, the evaluation plug-in obtaining sub-unit further includes:

[0055] The training set obtaining micro-unit is configured to divide the multiple sample monitoring data sets and sample risk coefficient sets as training data and divide them into K parts to obtain K training sets.

[0056] The evaluation model obtaining micro-unit is configured to utilize the K training sets to respectively supervise the training of the feedforward neural network until convergence to obtain K first risk evaluation models.

[0057] The evaluation plug-in construction micro-unit is configured to construct the first risk evaluation plug-in according to the K first risk evaluation models based on ensemble learning.

[0058] In the embodiments of the present application, the first risk evaluation model is a model obtained by respectively supervising the training of the feedforward neural network until convergence using the K training sets obtained by dividing the multiple sample monitoring data sets and sample risk coefficient sets into K parts.

[0059] Specifically, the training set obtaining micro-unit first plays a role. It takes the multiple sample monitoring data sets and sample risk coefficient sets obtained based on the first public pipeline as training data. To ensure the reliability and generalization ability of model training, these data are divided into K parts to obtain K training sets. This operation can fully utilize the data and avoid problems such as uneven data distribution and overfitting affecting the performance of the model.

[0060] Then, the evaluation model uses the K training sets to supervise the training of the feedforward neural network. During the training process, each training set is input into the feedforward neural network, and the network calculates the predicted risk evaluation results according to the input data. Then the predicted results are compared with the actual sample risk coefficients in the training set, and the difference between the two is calculated through the back propagation algorithm, and the weights and thresholds of the network are adjusted according to the difference to optimize the performance of the network. This iteration is repeated until the network converges, and finally K first risk evaluation models are obtained. Each model has learned the relationship between sample monitoring data and risk coefficients from different training data subsets, and has its own advantages.

[0061] Finally, the evaluation plug-in constructs a micro-unit based on the method of ensemble learning to integrate the K first risk evaluation models to construct the first risk evaluation plug-in. First, the K first risk evaluation models are regarded as independent individuals. These models are trained using different training sets on the feedforward neural network, and each has learned part of the relationship between sample monitoring data and risk coefficients. Then, according to the specific ensemble strategy, such as weighted average of the output results of the K models, the weights are allocated according to the performance of each model in training, and the model with better performance has higher weight; or use the voting method, let K models evaluate the risk of the same set of monitoring data, and determine the final risk evaluation coefficient according to the evaluation results of the majority of models. In this way, the advantages of the K models are combined to finally construct the first risk evaluation plug-in that can more accurately and stably evaluate the running risk of the first public pipeline.

[0062] By fusing the results of the K models, the first risk evaluation plug-in constructed can more accurately evaluate the running risk of the first public pipeline, overcoming the limitations of traditional single models, and providing more reliable support for risk prevention and control of underground comprehensive pipe galleries.

[0063] In one possible implementation, the comprehensive risk calculation module 30 further includes:

[0064] The monitoring record running unit is configured to count the frequency and scale of each public pipeline failure according to the pipe gallery operation monitoring records within a preset time range, and obtain a plurality of failure frequencies and a plurality of failure scales.

[0065] The weight determination unit is configured to determine a plurality of weight proportions based on the plurality of failure frequencies and the plurality of failure scales.

[0066] The risk calculation unit is configured to perform comprehensive risk calculation on the plurality of risk coefficients according to the plurality of weight proportions, and obtain an overall risk coefficient.

[0067] In the embodiments of the present application, the fault frequency is obtained by counting the number of faults of each public pipeline according to the pipeline operation monitoring records within a preset time range. The fault scale is obtained by counting the range of influence and the size of loss caused by the fault of each public pipeline according to the pipeline operation monitoring records within a preset time range. The overall risk coefficient is obtained by calculating the comprehensive risk of multiple risk coefficients based on the fault frequency and scale of each public pipeline counted from the pipeline operation monitoring records within a preset time range, and determining the weight proportion.

[0068] Specifically, first, the monitoring record operation unit works. It collects pipeline operation monitoring records within a preset time range, which details the operation of power, communication, water supply pipeline and other public pipelines at different times. Through in-depth analysis of these records, the frequency and scale of faults of each public pipeline are counted. For example, the number of power pipeline fault occurrences in the past year, and the regional range or loss size affected by each fault; for communication pipelines, the frequency of signal interruption and the number of users affected by interruption are counted, thereby obtaining multiple fault frequency and multiple fault scale data.

[0069] Then, the weight determination unit works. It performs comprehensive analysis based on the multiple fault frequency and multiple fault scale data obtained by the monitoring record operation unit. The higher the fault frequency and the larger the scale of the public pipeline, the higher the weight it should occupy in the overall risk assessment, because its impact on the safety of pipeline operation is more significant. Through the analytic hierarchy process, the weight proportion corresponding to each public pipeline is determined, and the sum of multiple weight proportions is ensured to be 1, ensuring the rationality and scientificity of weight distribution.

[0070] First, a hierarchical structure model is constructed, taking the overall risk of the pipeline as the target layer; taking the fault frequency and fault scale of power, communication, water supply pipeline and other public pipelines as the criterion layer; and taking each type of public pipeline itself as the scheme layer. The relative importance of each element in the criterion layer and the scheme layer is determined by pairwise comparison to construct a judgment matrix. The largest eigenvalue and eigenvector of the judgment matrix are calculated using the eigenvalue method or other methods, and then the relative weight of each element to the upper element is obtained. After obtaining the weight of each public pipeline based on the fault frequency and fault scale, the multiple weight proportions are normalized to ensure that the sum of the multiple weight proportions is 1. In this way, both the influence of fault frequency and scale of different public pipelines on the overall risk and the rationality and scientificity of weight distribution ensured by rigorous mathematical calculation are considered.

[0071] Finally, the risk calculation unit determines the overall risk coefficient of the underground comprehensive pipe gallery based on the multiple weight proportions obtained by the weight determination unit and the multiple risk coefficients obtained by the operation risk assessment module 20. The risk coefficient of each public pipeline is multiplied by its corresponding weight proportion, and all the products are added to obtain the overall risk coefficient that can comprehensively reflect the overall operation risk of the underground comprehensive pipe gallery. This coefficient comprehensively considers the risk of multiple public pipelines, and compared with the prior art, can more accurately assess the risk of the pipe gallery and provide a reliable basis for subsequent risk warning and prevention.

[0072] In one possible implementation manner, the weight determination unit further includes:

[0073] The weight proportion is positively correlated with the failure frequency and the failure scale, and the sum of the multiple weight proportions is 1.

[0074] Specifically, first, the operation monitoring records of the pipe gallery are obtained. By collecting detailed records in a preset time range, the failure frequency and scale of various public pipelines such as power, communication, and water supply pipelines are counted, which are the basic data for subsequent weight determination. For example, the number of power pipeline failures in the past year, the size of the power outage area caused by each failure, the number of affected users, and other scale data are counted; and the number of signal interruptions of the communication pipeline and the communication range affected by each interruption are counted.

[0075] Then, the weight analysis is performed using the failure frequency and scale data. Since the weight proportion is positively correlated with the failure frequency and the failure scale, the more frequent and larger the failure, the more important the pipeline is in the overall risk assessment, and the greater the corresponding weight proportion. For example, if the gas pipeline has a high failure frequency in the statistical period, and each failure causes a wide range of danger and a high degree of harm, then its weight proportion in the comprehensive risk calculation will be relatively large.

[0076] In order to ensure the scientificity of the weight distribution, the analytic hierarchy process (AHP) is adopted. A hierarchical structure model is constructed, the overall risk of the pipe gallery is taken as the target layer, the failure frequency and the failure scale of each type of public pipeline are taken as the criterion layer, and each type of public pipeline is taken as the scheme layer. Through expert scoring or data analysis, the elements of the criterion layer and the scheme layer are compared two by two to construct a judgment matrix. Then, the maximum eigenvalue and the eigenvector of the judgment matrix are calculated by using a professional algorithm, so as to obtain the relative weight of each public pipeline with respect to the overall risk.

[0077] Finally, in order to ensure the normalization and additivity of the weights, the calculated weight proportions are normalized. Through specific mathematical operations, the sum of the weight proportions of all public pipelines is 1. In this way, the mathematical logic requirements are met, and each weight proportion is within a reasonable range, accurately reflecting the relative importance of each public pipeline in the comprehensive risk calculation, providing a scientific basis for subsequent accurate calculation of the overall risk coefficient based on multiple risk coefficients.

[0078] In one possible implementation, the operation risk warning module 40 further includes:

[0079] If any of the risk coefficients is greater than the predetermined risk scalar, a pipeline risk warning signal for the corresponding public pipeline is generated, and a pipeline operation risk warning is performed.

[0080] Specifically, first, the operation risk of each public pipeline in the pipe gallery is evaluated based on the multi-source monitoring data set, and multiple risk coefficients are obtained. These risk coefficients are derived based on the analysis of real-time pipeline operation data and can reflect the current risk level of the pipeline.

[0081] Then, each risk coefficient is compared with a predetermined risk scalar. The predetermined risk scalar is a key indicator determined based on various factors such as the safe operation standards of various public pipelines in the pipe gallery, historical failure data, and actual risk prevention and control needs, and is an important basis for determining whether the pipeline is in a risk state.

[0082] When any of the risk coefficients is found to be greater than the predetermined risk scalar, the warning signal generation unit generates a pipeline risk warning signal for the corresponding public pipeline. For example, if the risk coefficient of the power pipeline exceeds the predetermined risk scalar, a warning signal specific to the power pipeline is generated, which contains key information such as pipeline type and risk level.

[0083] Subsequently, the pipeline operation risk warning operation is performed. This may include various ways, such as issuing an audible and visual alarm in the monitoring center to attract the attention of staff; sending notifications to the terminal devices of relevant management personnel, such as mobile phones and computers, to ensure that they can obtain risk information in a timely manner; it may also be linked to other systems, such as starting emergency lighting and closing valves in relevant areas, to reduce the harm that risks may cause and ensure the safe operation of the pipe gallery and its pipelines.

[0084] Through this process, compared with the prior art, the risk problems of public pipelines in underground comprehensive pipe galleries can be more accurately and timely discovered and handled.

[0085] It should be noted that the above-mentioned embodiment sequences of the present application are merely for description only, but not for representing the advantages and disadvantages of the embodiments. And the above-mentioned embodiments of the present specification have been described. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0086] The above only describes the preferred embodiments of the present application, and does not limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0087] The specification and drawings are merely exemplary of the present application, and any and all modifications, variations, combinations or equivalents that are within the scope of the present application should be included. Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A risk prevention and control platform for underground utility tunnels, characterized in that, The risk prevention and control platform for the underground comprehensive pipe gallery comprises: A fixed-point monitoring module for fixed-point monitoring of multiple public pipelines in the underground comprehensive pipe gallery by using the Internet of Things to obtain a multi-source monitoring data set; An operation risk assessment module for performing operation risk assessment according to the multi-source monitoring data set to obtain multiple risk coefficients; A comprehensive risk calculation module for performing comprehensive risk calculation based on the multiple risk coefficients to obtain an overall risk coefficient; An operation risk early warning module for performing pipe gallery operation risk early warning if the overall risk coefficient is greater than a predetermined risk coefficient threshold; The multiple public pipelines at least include power pipelines, communication pipelines, water supply pipelines, drainage pipelines, gas pipelines and heat supply pipelines; The comprehensive risk calculation module comprises: A monitoring record operation unit for respectively counting the failure frequency and scale of each public pipeline according to pipe gallery operation monitoring records within a preset time range to obtain multiple failure frequencies and multiple failure scales; A weight determination unit for determining multiple weight proportions based on the multiple failure frequencies and multiple failure scales; A risk calculation unit for performing comprehensive risk calculation on the multiple risk coefficients according to the multiple weight proportions to obtain an overall risk coefficient; The weight proportions are positively correlated with the failure frequency and the failure scale, and the sum of the multiple weight proportions is 1; If any risk coefficient in the multiple risk coefficients is greater than a predetermined risk scalar, a pipeline risk early warning signal of the corresponding public pipeline is generated, and pipeline operation risk early warning is performed; The operation risk assessment module comprises: A pipeline selection unit for randomly selecting one public pipeline as a first public pipeline from the multiple public pipelines; A plug-in evaluation unit for constructing a first risk assessment plug-in based on the first public pipeline; A risk coefficient adding unit for obtaining a first monitoring data set of the first public pipeline, inputting the first risk assessment plug-in, outputting a first risk assessment coefficient and adding the first risk assessment coefficient to the multiple risk coefficients; The plug-in evaluation unit comprises: An index set monitoring subunit for obtaining a first monitoring index set of the first public pipeline; An index set constraint subunit for taking the first monitoring index set as a constraint, retrieving a historical operation log of the first public pipeline, obtaining multiple sample monitoring data sets, and analyzing pipeline failure risks under different sample monitoring data to obtain a sample risk coefficient set; An evaluation plug-in obtaining subunit for training a feedforward neural network to convergence by using the multiple sample monitoring data sets and the sample risk coefficient set to obtain the first risk assessment plug-in; The evaluation plug-in obtaining subunit comprises: A training set obtaining micro unit for taking the multiple sample monitoring data sets and the sample risk coefficient set as training data and equally dividing them into K parts to obtain K training sets; An evaluation model obtaining micro unit for respectively performing supervised training on a feedforward neural network by using the K training sets until convergence to obtain K first risk assessment models; An evaluation plug-in construction micro unit for constructing the first risk assessment plug-in according to the K first risk assessment models based on ensemble learning.

Images