Gas pipeline dynamic risk early warning platform and method based on Internet of Things

Abstract

The invention discloses a gas pipeline dynamic risk early warning platform and method based on the Internet of Things, and relates to the technical field of gas pipeline risk analysis, and the method comprises the following steps: obtaining first pressure and temperature data of a gas pipeline, carrying out the pressure error correction processing, and carrying out the gas leakage risk index calculation, and obtaining first gas leakage feature data; constructing a gas leakage risk quantification model, and calculating a gas leakage risk based on the gas leakage risk quantification model to obtain gas leakage risk data; performing gas leakage risk early warning on the gas pipeline based on the gas leakage risk data; the method is used for solving the problems that when an existing gas pipeline risk analysis technology carries out risk analysis on the gas leakage risk of a gas pipeline through gas pressure, pressure changes caused by tiny leakage cannot be accurately recognized in normal gas pressure fluctuations and pressure fluctuations caused by gas temperature changes; and meanwhile, the problem of gas leakage risk is analyzed.

Description

Technical Field

[0001] The present invention relates to the technical field of gas pipeline risk analysis, and in particular to a gas pipeline dynamic risk early warning platform and method based on the Internet of Things. Background Art

[0002] Gas pipeline risk analysis technology is a systematic technical means to evaluate the potential risks of gas pipeline systems. With the widespread use of gas in industrial production and daily life, the safety of gas pipelines is of vital importance. Once an accident occurs in a gas pipeline, it may lead to serious consequences such as fire and explosion. The purpose of risk analysis technology is to comprehensively inspect and evaluate gas pipelines, discover dangerous factors that may lead to accidents in advance, and take effective measures to prevent and control them.

[0003] When analyzing the gas leakage risk of a gas pipeline, the existing gas pipeline risk analysis technology usually analyzes the gas pressure data in the gas pipeline. However, when the gas pressure in the gas pipeline is collected by a pressure sensor, if the gas pressure in the pipeline itself has normal fluctuations, these normal fluctuations may not only mask the pressure changes caused by small leaks, but may also lead to misjudgment of gas leakage during risk analysis. For example, in the patent application with publication number CN118114885A, a dynamic risk early warning platform for gas pipelines based on the Internet of Things model is disclosed. When analyzing the gas leakage risk, this solution collects gas pressure data to Risk analysis is performed, but the gas pressure is still collected through conventional means, and the pressure changes caused by small leaks that are concealed by normal pressure fluctuations cannot be identified; and the gas pressure of the gas in the gas pipeline is also related to the temperature of the gas itself. If the gas pipeline is in an environment with obvious temperature changes, it is easy to misjudge the pressure fluctuations caused by temperature changes as gas leaks. Therefore, when the existing gas pipeline risk analysis technology performs risk analysis on the gas leakage risk of the gas pipeline through the gas pressure, it is unable to accurately identify the pressure changes caused by small leaks in the normal gas pressure fluctuations and the pressure fluctuations caused by the gas temperature changes, and analyze the gas leakage risk at the same time. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems in the prior art to a certain extent.

[0005] The first pressure and temperature data of the gas pipeline are obtained; the pressure error correction processing and the gas leakage risk index calculation are performed to obtain the first gas leakage characteristic data; a gas leakage risk quantification model is constructed to calculate the gas leakage risk and obtain the gas leakage risk data; a gas leakage risk warning is performed on the gas pipeline based on the gas leakage risk data; in order to solve the problem that the existing gas pipeline risk analysis technology cannot accurately identify the pressure changes caused by small leaks in the normal gas pressure fluctuations and the pressure fluctuations caused by the gas temperature changes when analyzing the gas leakage risk of the gas pipeline through the gas pressure, and at the same time analyze the gas leakage risk.

[0006] To achieve the above objectives, in the first aspect, the present application provides a gas pipeline dynamic risk warning platform based on the Internet of Things, including a data acquisition module, a data processing module, a risk quantification module and a dynamic warning module;

[0007] The data acquisition module is used to acquire first pressure and temperature data of the gas pipeline; the first pressure and temperature data includes first gas pressure data and first ambient temperature data;

[0008] The data processing module includes an error correction unit and a feature calculation unit. The error correction unit performs pressure error correction processing on the first gas pressure data based on the first ambient temperature data. The feature calculation unit is used to calculate the gas leakage risk index to obtain the first gas leakage feature data.

[0009] The risk quantification module includes a model building unit and a risk quantification unit. The model building unit is used to build a gas leakage risk quantification model. The risk quantification unit calculates the gas leakage risk based on the gas leakage risk quantification model to obtain gas leakage risk data.

[0010] The dynamic early warning module provides gas leakage risk early warning for gas pipelines based on gas leakage risk data.

[0011] Furthermore, the data acquisition module is configured with a data acquisition strategy, which includes:

[0012] A pressure sensor, a flow rate sensor and a temperature sensor are arranged at a first distance interval based on the gas pipeline. The pressure sensor, the flow rate sensor and the temperature sensor are arranged at the same position, and the first distance interval is L1; the pressure sensor, the flow rate sensor and the temperature sensor are marked as measurement sensors, and according to the order of the arrangement positions of the measurement sensors, the measurement sensor at any position is marked as A i , where i represents the i-th position, 1≤i.

[0013] Furthermore, the data acquisition module is configured with a data acquisition strategy, and the data acquisition strategy also includes:

[0014] For measuring sensor A i , when i=1, the gas pressure, gas flow rate and gas temperature in the gas pipeline are obtained at a first time interval, and the acquisition time and the number of acquisitions are recorded. The first time interval is t1;

[0015] When i=2, the first arrival time is calculated according to the flow rate of the gas obtained by A2 each time and the first distance interval L1, and according to the first arrival time and the acquisition time recorded by A1, the gas pressure, gas flow rate and gas temperature in the gas pipeline are acquired after the first arrival time length is delayed after each acquisition by A1, and the acquisition time and the number of acquisitions are recorded;

[0016] When i>2, according to A i-1 and A i-2 The second arrival time is calculated based on the gas flow rate obtained each time, the recorded acquisition time and the first distance interval L1, and the second arrival time and A are calculated based on the second arrival time and A. i-1 The acquisition time of the record is in A i-1 After each acquisition, the gas pressure, gas flow rate and gas temperature are acquired after a delay of the second arrival time, and the acquisition time and the number of acquisitions are recorded;

[0017] The gas pressures, corresponding acquisition times, and acquisition times acquired by all measuring sensors are marked as first gas pressure data;

[0018] Marking the ambient temperature acquired by the temperature sensors corresponding to all the measuring sensors and the corresponding acquisition time and acquisition number as first ambient temperature data;

[0019] Gas leakage in the gas pipeline is recorded as M=1, and no gas leakage in the gas pipeline is recorded as M=0. The gas pipeline is marked according to whether leakage occurs when the first gas pressure data and the first pressure and temperature data are obtained; and the corresponding measurement sensors are combined and stored and recorded as the first pressure and temperature data.

[0020] Furthermore, the error correction unit is configured with an error correction strategy, and the error correction strategy includes:

[0021] Set the standard gas temperature to K0; based on the first ambient temperature data and the first gas pressure data; according to the gas pressure and gas temperature at the same acquisition time, use the pressure correction formula to calculate the gas pressure at the standard gas temperature, recorded as the standard gas pressure, the pressure correction formula is as follows: Wherein Pe represents the gas pressure at the standard gas temperature, Pa represents the obtained gas pressure, and Ka represents the gas temperature corresponding to the obtained gas pressure; and the corresponding gas pressure is replaced with the calculated standard gas pressure, and the second gas pressure data is obtained after completion.

[0022] Furthermore, the feature calculation unit is configured with a feature calculation strategy, and the feature calculation strategy includes:

[0023] For any two adjacent positions on the gas pipeline, set as i and i+1, the gas leakage risk index is calculated according to the corresponding second gas pressure data, and the first gas leakage characteristic data is obtained after completion; the gas leakage risk index calculation includes calculating the pressure difference characteristic, calculating the pressure difference change rate characteristic, calculating the pressure difference gradient characteristic and calculating the pressure difference fluctuation characteristic;

[0024] Calculating the pressure difference feature includes: assuming that the second gas pressure data corresponding to the pressure sensors at any two adjacent positions on the gas pipeline are P i，j and P i+1，j , where j represents the number of acquisitions; based on P i，j and P i+1，j The pressure difference characteristic is calculated using the pressure difference calculation formula. The pressure difference calculation formula is as follows: B i,j =|P i,j -P i+1,j |, where B i，j Represents the pressure difference characteristics between the i-th position and the i+1-th position on the gas pipeline at the j-th acquisition;

[0025] Calculating the pressure difference change rate characteristics includes: assuming P i，j The corresponding acquisition time is t i，j , P i，j-1 The acquisition time is t i，j-1 , the pressure difference change rate characteristic is calculated using the pressure difference change rate calculation formula. The pressure difference change rate calculation formula is as follows: Among them C i，j Representative B i，j With B i，j+1 The pressure difference change rate characteristic, B i，j+1 Represents the pressure difference characteristics between the i-th position and the i+1-th position on the gas pipeline during the j-1-th acquisition.

[0026] Furthermore, the feature calculation unit is configured with a feature calculation strategy, and the feature calculation strategy further includes:

[0027] Calculating the pressure difference gradient feature includes: based on the pressure difference feature B i，j And the first distance interval L1, the pressure difference gradient characteristic is calculated using the pressure difference gradient calculation formula, and the pressure difference gradient calculation formula is as follows: Where D i，j Representative B i，j Pressure gradient characteristics;

[0028] Calculating the pressure difference fluctuation characteristics includes: setting the first time length as T0, obtaining the pressure difference characteristics within the first time length, and sorting them according to the corresponding time series, marking them as the first pressure difference characteristic sequence, and calculating the pressure difference fluctuation characteristics within the first time length using the pressure difference fluctuation calculation formula. The pressure difference fluctuation calculation formula is as follows: Where E i，j represents the pressure difference fluctuation feature of the i-th position within the corresponding first time length with the j-th acquisition corresponding time as the end point, Bn represents the n-th pressure difference feature in the first pressure difference feature sequence, B0 represents the average value of the first pressure difference feature sequence; N represents the total number of pressure difference features in the first pressure difference feature sequence;

[0029] The pressure difference characteristics, pressure difference change rate characteristics, pressure difference gradient characteristics and pressure difference fluctuation characteristics are marked as gas leakage risk indicators.

[0030] Furthermore, the model building unit is configured with a model building strategy, and the model building strategy includes:

[0031] A leakage risk logistic regression model is constructed based on the logistic regression model. The leakage risk logistic regression model is as follows: Where Y(M=1) represents the risk probability of leakage; B, C, D and E represent the pressure difference characteristic, the pressure difference change rate characteristic, the pressure difference gradient characteristic and the pressure difference fluctuation characteristic in order, respectively; β0, β1, β2, β3 and β4 are the parameters of the leakage risk logistic regression model;

[0032] The learning rate of model training is set to h1, the training round is set to h2, and the batch size is set to h3. The leakage risk logistic regression model is trained using the first gas leakage feature data to obtain a gas leakage risk quantification model.

[0033] Furthermore, the risk quantification unit is configured with a risk quantification strategy, which includes:

[0034] The most recently acquired first gas leakage characteristic data is input into the gas leakage risk quantification model at a second time interval to obtain the gas leakage risk of all corresponding positions on the gas pipeline, which is marked as gas leakage risk data. The second time interval is t2, t2>t1, and the gas leakage risk range is [0, 1].

[0035] Furthermore, the dynamic warning module is configured with a dynamic warning strategy, which includes:

[0036] Set the risk threshold to Y0, and based on the gas leakage risk data, determine whether any gas leakage risk satisfies or exceeds Y0; if it satisfies or exceeds Y0, trigger an alarm signal, and mark the position on the gas pipeline corresponding to the gas leakage risk greater than or equal to Y0 as a dangerous position, and also mark the position on the gas pipeline next to the dangerous position as a dangerous position, and mark the gas pipeline between the two dangerous positions as a leaking pipeline;

[0037] If it does not meet the requirement of being greater than or equal to Y0, for the gas leakage risk corresponding to any position on the gas pipeline, determine whether it meets the requirement of being greater than or equal to 0.85*Y0 and increases for N1 consecutive times. If it meets the requirement, the early warning signal is triggered, and the corresponding position on the gas pipeline is marked as a risk position, and the position after the risk position on the gas pipeline is also marked as a risk position, and the gas pipeline between the two risk positions is marked as a risk pipeline; if it does not meet the requirement, the corresponding position on the gas pipeline is marked as non-risk.

[0038] In the second aspect, the present application provides a gas pipeline dynamic risk early warning method based on the Internet of Things, comprising the following steps:

[0039] Acquire first pressure and temperature data of the gas pipeline; the first pressure and temperature data includes first gas pressure data and first ambient temperature data;

[0040] Performing pressure error correction processing on the first gas pressure data based on the first ambient temperature data, and calculating the gas leakage risk index to obtain first gas leakage characteristic data;

[0041] Construct a gas leakage risk quantification model, and calculate the gas leakage risk based on the gas leakage risk quantification model to obtain gas leakage risk data;

[0042] Provide gas leakage risk warning for gas pipelines based on gas leakage risk data.

[0043] Beneficial effects of the present invention: The present invention obtains first pressure and temperature data of the gas pipeline; the first pressure and temperature data includes first gas pressure data and first ambient temperature data; based on the first ambient temperature data, the first gas pressure data is subjected to pressure error correction processing, and a gas leakage risk index is calculated to obtain first gas leakage characteristic data; a gas leakage risk quantification model is constructed, and the gas leakage risk is calculated based on the gas leakage risk quantification model to obtain gas leakage risk data; based on the gas leakage risk data, a gas leakage risk warning is performed on the gas pipeline; in normal gas pressure fluctuations and pressure fluctuations caused by gas temperature changes, the pressure changes caused by small leaks can be accurately identified, and then the gas leakage risk can be analyzed;

[0044] The present invention adjusts the gas pressure collection time through the gas flow rate to ensure that the pressure sensor collects the pressure of part of the gas in the same collection. The advantage is that more accurate pressure change data can be obtained, which avoids misjudgment caused by collecting the pressures of different parts of the gas during normal fluctuations in the gas pressure. The gas pressure is corrected by the gas temperature to avoid the pressure change caused by temperature change being misjudged as leakage risk, thereby improving the accuracy and consistency of risk analysis. By calculating a variety of characteristic data to perform gas leakage risk analysis, the status of the gas pipeline can be analyzed from multiple dimensions, abnormal conditions in the operation of the gas pipeline can be captured more comprehensively, and small leakage risks can be discovered more effectively. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 is a functional block diagram of the system of the present invention;

[0046] Figure 2 is a flow chart of the steps of the method of the present invention;

[0047] Figure 3 It is a schematic diagram of setting the gas pipeline sensor of the present invention;

[0048] Figure 4 It is a flow chart of the dynamic early warning strategy of the present invention;

[0049] Figure 5 It is a schematic structural diagram of the electronic device of the present invention. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] Example 1, please refer to Figure 1 As shown, the present application provides a gas pipeline dynamic risk warning platform based on the Internet of Things, including a data acquisition module, a data processing module, a risk quantification module and a dynamic warning module;

[0052] The data acquisition module is used to acquire first pressure and temperature data of the gas pipeline; the first pressure and temperature data includes first gas pressure data and first ambient temperature data;

[0053] The data acquisition module is configured with a data acquisition strategy, which includes: Figure 3As shown, based on the gas pipeline, a pressure sensor, a flow rate sensor and a temperature sensor are arranged at a first distance interval. The pressure sensor, the flow rate sensor and the temperature sensor are arranged at the same position, that is, three sensors are placed at the same position on the gas pipeline. The first distance interval is L1. In this embodiment, the first distance interval L1 is 500 meters; the pressure sensor, the flow rate sensor and the temperature sensor are marked as measurement sensors. According to the arrangement position sequence of the measurement sensors, the measurement sensor at any position is marked as A. i , where i represents the i-th position, 1≤i, that is, the starting position number is 1;

[0054] For measuring sensor A i , when i=1, that is, the measuring sensor A1 at the first position of the gas pipeline obtains the gas pressure, gas flow rate and gas temperature in the gas pipeline at a first time interval, and records the acquisition time and the number of acquisitions. The first time interval is t1; in this embodiment, the first time interval is 10 seconds;

[0055] When i=2, i.e., the measuring sensor A2 at the second position of the gas pipeline calculates the first arrival time according to the gas flow rate obtained by A2 each time and the first distance interval L1, and calculates according to time=distance / flow rate; and according to the first arrival time and the acquisition time recorded by A1, after each acquisition by A1, the gas pressure, gas flow rate and gas temperature in the gas pipeline are acquired after the first arrival time is delayed, and the acquisition time and the number of acquisitions are recorded; for example, the first distance interval L1=500 meters, the first acquisition time of A1 is 14:10:10 seconds, and the collected gas flow rate is 10m / s, then the first arrival time=500 / 10=50 seconds, and the first acquisition time of A2 is 14:10:10 seconds+50 seconds=14:11:0 seconds;

[0056] When i>2, that is, the measuring sensor after the second position of the gas pipeline, according to A i-1 and A i-2 The second arrival time is calculated by using the gas flow rate obtained each time, the recorded acquisition time and the first distance interval L1. First, according to A i-1 and A i-2 The obtained gas flow rate and the recorded acquisition time are used to calculate the acceleration, and then the second arrival time is solved according to the relationship between acceleration and distance; and the second arrival time and A are used to calculate the acceleration. i-1 The acquisition time of the record is in A i-1 After each acquisition, the gas pressure, gas flow rate and gas temperature are acquired after a delay of the second arrival time, and the acquisition time and the number of acquisitions are recorded; for example, the first distance interval L1 = 500 meters, A i-2 The second collection time is 14:10:10, the collected flow velocity is 10m / s, Ai-1 The second collection time is 14:11:00, and the collected flow velocity is 11m / s. According to 11^2-10^2=2*acceleration*L1, we get acceleration=0.021m / s^2. According to L1=11*t+(acceleration*t^2) / 2, we solve t=43.6 seconds, that is, the second arrival time=43.6 seconds, that is, A i The second collection time is 14:11:00+43.6=14:11:43.6;

[0057] The gas pressures, corresponding acquisition times, and acquisition times acquired by all measuring sensors are marked as first gas pressure data;

[0058] Marking the ambient temperature acquired by the temperature sensors corresponding to all the measuring sensors and the corresponding acquisition time and acquisition number as first ambient temperature data;

[0059] The occurrence of gas leakage in the gas pipeline is recorded as M=1, and the occurrence of no gas leakage in the gas pipeline is recorded as M=0, and the gas pipeline is marked according to whether the first gas pressure data and the first pressure and temperature data are obtained. The data are combined and stored according to the corresponding measurement sensors and recorded as the first pressure and temperature data. For the pressure data and temperature data when the gas pipeline leaks, if they cannot be obtained in reality, they can also be collected through laboratory simulation. By establishing a simulation model of the gas pipeline, a small leak is artificially caused in the simulation model, and then the pressure data and temperature data at the time of the leak are recorded;

[0060] In a specific embodiment, by adjusting the collection time interval to ensure that the two pressure sensors collect the pressure of the same part of the gas, more accurate pressure change data can be obtained; accurate pressure data is crucial for the identification of small gas leakage risks; because small leaks may only cause very slight pressure changes; for example, assuming that there is no such time interval adjustment, the two pressure sensors may collect the pressures of different parts of the gas. When the gas pressure in the pipeline itself has normal fluctuations, these unsynchronized data may mask the pressure changes caused by small leaks, making data analysis complicated and inaccurate. The adjustment time is calculated by treating the gas as uniformly accelerated motion because the gas has high fluidity as a gas. For pressure collection, it is only necessary to ensure that the collected pressure comes from roughly the same part of the gas. Approximating it as uniformly accelerated motion can greatly simplify the calculation process when the accuracy requirement is not high, and it can also be calculated using other methods according to actual needs.

[0061] The data processing module includes an error correction unit and a feature calculation unit. The error correction unit performs pressure error correction processing on the first gas pressure data based on the first ambient temperature data. The feature calculation unit is used to calculate the gas leakage risk index to obtain the first gas leakage feature data.

[0062] The error correction unit is configured with an error correction strategy, which includes: setting the standard gas temperature as K0; a reference temperature for correcting gas at different temperatures, in Kelvin. In this embodiment, the standard gas temperature K0 = 293.15 Kelvin, i.e., 20 degrees Celsius;

[0063] Based on the first ambient temperature data and the first gas pressure data, and according to the gas pressure and gas temperature at the same acquisition time, the gas pressure at the standard gas temperature is calculated using a pressure correction formula and recorded as the standard gas pressure. The pressure correction formula is as follows: Where Pe represents the gas pressure at the standard gas temperature, Pa represents the obtained gas pressure, and Ka represents the gas temperature corresponding to the obtained gas pressure; and the corresponding gas pressure is replaced with the calculated standard gas pressure, and the second gas pressure data is obtained after completion; for example, a measurement sensor at a certain position obtains that when the gas temperature is 303.15 Kelvin, the gas pressure is 650.00 kpa, then the gas pressure at the standard gas temperature K0 = 293.15 Kelvin is 650*293.15 / 303.15 = 628.56 kpa;

[0064] The characteristic calculation unit is configured with a characteristic calculation strategy, which includes: for any two adjacent positions on the gas pipeline, set as i and i+1, a gas leakage risk index is calculated according to the corresponding second gas pressure data, and the first gas leakage characteristic data is obtained after completion; the gas leakage risk index calculation includes calculating a pressure difference characteristic, calculating a pressure difference change rate characteristic, calculating a pressure difference gradient characteristic, and calculating a pressure difference fluctuation characteristic;

[0065] Calculating the pressure difference feature includes: assuming that the second gas pressure data corresponding to the pressure sensors at any two adjacent positions on the gas pipeline are P i，j and P i+1，j , that is, the pressure data after error correction; where j represents the number of acquisitions; based on P i，j and P i+1，j The pressure difference characteristic is calculated using the pressure difference calculation formula. The pressure difference calculation formula is as follows: B i,j =|P i,j -P i+1,j |, where B i，jrepresents the pressure difference characteristic between the i-th position and the i+1-th position on the gas pipeline at the j-th acquisition; for example, the pressure sensors at the second and third positions acquire the corresponding P 2，3 and P 3，3 P 2，3 =650kpa, P 3，3 =640kpa, then B 2，3 =|650-640|=10kpa; Under normal circumstances, when the same part of gas flows through adjacent sensors, the pressure difference should be within a relatively stable range; if there is a small leak, the pressure near the leak point will drop, causing the pressure difference to change;

[0066] Calculating the pressure difference change rate characteristics includes: assuming P i，j The corresponding acquisition time is t i，j , P i，j-1 The acquisition time is t i，j-1 , the pressure difference change rate characteristic is calculated using the pressure difference change rate calculation formula. The pressure difference change rate calculation formula is as follows: Among them C i，j Representative B i，j With B i，j+1 The pressure difference change rate characteristic, B i，j+1 represents the pressure difference characteristic between the i-th position and the i+1-th position on the gas pipeline at the j-1th acquisition; when calculating, for example, B 2，2 =10.35kpa, B 2，1 =10.29kpa,t 2，1 = 15 hours 10 minutes 50 seconds, t 2，1 = 15 hours 10 minutes 38 seconds, then C 2，2 =(10.35-10.29) / (50-38)=0.0059kpa / s; The pressure difference change rate feature is used to measure the dynamic change of the pressure difference; even if the pressure difference is within the normal range at a certain moment, if its change rate is abnormal, it may also indicate a small leakage risk; for example, when the gas starts to leak, the pressure difference may gradually increase, causing the pressure difference change rate to be non-zero and beyond the normal fluctuation range;

[0067] Calculating the pressure difference gradient feature includes: based on the pressure difference feature B i，j And the first distance interval L1, the pressure difference gradient characteristic is calculated using the pressure difference gradient calculation formula, and the pressure difference gradient calculation formula is as follows: Where D i，j Representative B i，j The pressure gradient characteristics; for example, B 2，2 =10.35kpa, L1 = 500m, then D 2，2=10.35 / 500=0.0207kpa / m; Abnormal changes in pressure gradient characteristics may indicate local resistance changes or leaks in the pipeline, because leaks can cause local pressure imbalances and cause the pressure gradient to deviate from the normal range;

[0068] Calculating the pressure difference fluctuation characteristics includes: setting the first time length as T0, in this embodiment, T0 = 30 minutes, obtaining the pressure difference characteristics within the first time length, and sorting them according to the corresponding time series, marked as the first pressure difference characteristic sequence, and calculating the pressure difference fluctuation characteristics within the first time length using the pressure difference fluctuation calculation formula, the pressure difference fluctuation calculation formula is as follows: Where E i，j represents the pressure difference fluctuation characteristics of the i-th position within the corresponding first time length with the j-th acquisition corresponding time as the end point, for example, T0 = 10 minutes, E 2，70 represents the second position, with the 70th acquisition time as the end point, and the pressure difference characteristics corresponding to the pressure difference fluctuation characteristics within the next 10 minutes. If a total of 60 times are acquired within 10 minutes, then the first pressure difference characteristic sequence is B 11 To B 70 ; Bn represents the nth pressure difference feature in the first pressure difference feature sequence, B0 represents the average value of the first pressure difference feature sequence; N represents the total number of pressure difference features in the first pressure difference feature sequence; the pressure difference fluctuation feature is used to measure the fluctuation of the pressure difference; under normal circumstances, the fluctuation of the pressure difference should be within a small range; if there is a small leak, it may cause irregular fluctuation of the pressure difference, resulting in an increase in the fluctuation of the pressure difference;

[0069] The pressure difference characteristics, pressure difference change rate characteristics, pressure difference gradient characteristics and pressure difference fluctuation characteristics are marked as gas leakage risk indicators;

[0070] In the specific implementation process, because the data is collected by adjusting the collection time interval to ensure that the two pressure sensors collect the pressure of the same part of the gas, the pressure sensors at different positions have different acquisition times for the same number of acquisitions. For example, the pressure sensors at the first position and the second position have the same number of acquisitions but differ by the first arrival time.

[0071] The risk quantification module includes a model building unit and a risk quantification unit. The model building unit is used to build a gas leakage risk quantification model. The risk quantification unit calculates the gas leakage risk based on the gas leakage risk quantification model to obtain gas leakage risk data.

[0072] The model building unit is configured with a model building strategy, which includes: building a leakage risk logistic regression model based on the logistic regression model, and the leakage risk logistic regression model is as follows: Where Y(M=1) represents the risk probability of leakage; B, C, D and E represent the pressure difference feature, the pressure difference change rate feature, the pressure difference gradient feature and the pressure difference fluctuation feature in order, respectively; β0, β1, β2, β3 and β4 are the parameters of the leakage risk logistic regression model; β0, β1, β2, β3 and β4 should be initialized before model training, and β0, β1, β2, β3 and β4 are usually initialized to small random values, for example, 0.005, because if the initial parameter value is too large, it will affect the training of the model; the logistic regression model is suitable for binary classification problems, such as the judgment of gas leakage risk, leakage risk and no leakage risk; it can learn according to multiple risk indicators and output a probability value in the interval, indicating the risk level of gas leakage; this probability output can provide more detailed information for risk assessment, which is convenient for operators to take different levels of response measures according to the risk level;

[0073] Set the learning rate of model training to h1, the training round to h2, and the batch size to h3. In this embodiment, h1=0.001, h2=200, and h3=32. Use the first gas leakage feature data to train the leakage risk logistic regression model to obtain a gas leakage risk quantification model.

[0074] The risk quantification unit is configured with a risk quantification strategy, which includes: inputting the newly acquired first gas leakage characteristic data into the gas leakage risk quantification model at a second time interval to obtain the gas leakage risk of all corresponding positions on the gas pipeline, marking it as gas leakage risk data, the second time interval is t2, t2>t1, and the gas leakage risk range is [0, 1]; the second time interval is 1 minute, that is, the gas pipeline is detected once every minute;

[0075] In the specific implementation process, when training the leakage risk logistic regression model, the first gas leakage feature data can be divided into a training set and a test set in advance, and the training set is used for training, and then the model performance is tested with the test set until the trained model performance meets the actual requirements.

[0076] The dynamic early warning module provides early warning of gas leakage risks for gas pipelines based on gas leakage risk data;

[0077] The dynamic warning module is configured with dynamic warning strategies, which include: Figure 4As shown, the risk threshold is set to Y0, and in this embodiment, Y0=0.60; the gas leakage risk data is traversed to determine whether a gas leakage risk satisfies or exceeds Y0; if it satisfies or exceeds Y0, an alarm signal is triggered, and the position on the gas pipeline corresponding to the gas leakage risk greater than or equal to Y0 is marked as a dangerous position, and the position on the gas pipeline next to the dangerous position is also marked as a dangerous position, and the gas pipeline between the two dangerous positions is marked as a leaking pipeline; for the judgment that the gas leakage risk is greater than or equal to the risk threshold, each judgment only needs to be made on the latest gas leakage risk data, and there is no need to make a judgment on the historical data that has been judged;

[0078] If it does not satisfy the condition of being greater than or equal to Y0, for the gas leakage risk corresponding to any position on the gas pipeline, determine whether it satisfies the condition of being greater than or equal to 0.85*Y0 and increases for N1 consecutive times. In the present embodiment, N1=4. If it satisfies the condition, a warning signal is triggered, and the corresponding position on the gas pipeline is marked as a risk position, and the position on the gas pipeline next to the risk position is also marked as a risk position, and the gas pipeline between the two risk positions is marked as a risk pipeline; for example, Y0=0.6, N1=4, and the gas leakage risks obtained at the third position in the most recent 4 times are 0.41, 0.43, 0.49 and 0.52 in chronological order from far to near, and increase for 4 consecutive times, and the most recent 0.52>0.85*0.60, so the third position is judged to be a risk position, and the gas pipeline between the third position and the fourth position is marked as a risk pipeline; if it does not satisfy the condition, the corresponding position on the gas pipeline is marked as risk-free.

[0079] In a specific embodiment, the risk threshold Y0 can be set according to the actual application scenario, and should not be too large or too small. If the threshold is set too high, for example, greater than 0.8, the model will be very cautious in judging that there is a leakage risk, and will only issue an early warning when the output probability is very high; although it will result in a high accuracy rate, many situations where there is actually a leakage risk may be missed because the probability does not reach this high threshold, resulting in an increased possibility of missed reports; when the threshold is set too small, for example, less than 0.4, the model will easily judge that there is a leakage risk, although it can capture almost all possible leakage situations; however, the accuracy will be very low, and a large number of false alarms will be generated; in actual applications, this will cause operation and maintenance personnel to frequently receive false early warning information, requiring a large number of unnecessary inspections and repairs on the pipeline, increasing operating costs.

[0080] Example 2, please refer to Figure 2 As shown, the present application provides a gas pipeline dynamic risk early warning method based on the Internet of Things, comprising the following steps:

[0081] Step S1, obtaining first pressure and temperature data of the gas pipeline; the first pressure and temperature data includes first gas pressure data and first ambient temperature data; Step S1 includes the following sub-steps:

[0082] Step S101, based on the gas pipeline, a pressure sensor, a flow rate sensor and a temperature sensor are arranged at a first distance interval, the pressure sensor, the flow rate sensor and the temperature sensor are arranged at the same position, and the first distance interval is L1; the pressure sensor, the flow rate sensor and the temperature sensor are marked as measurement sensors, and according to the arrangement position sequence of the measurement sensors, the measurement sensor at any position is marked as A i , where i represents the i-th position, 1≤i;

[0083] Step S102: for measuring sensor A i , when i=1, the gas pressure, gas flow rate and gas temperature in the gas pipeline are obtained at a first time interval, and the acquisition time and the number of acquisitions are recorded. The first time interval is t1;

[0084] Step S103, when i=2, the first arrival time is calculated according to the flow rate of the gas obtained by A2 each time and the first distance interval L1, and according to the first arrival time and the acquisition time recorded by A1, the gas pressure, gas flow rate and gas temperature in the gas pipeline are acquired after the first arrival time length is delayed after each acquisition by A1, and the acquisition time and the number of acquisitions are recorded;

[0085] Step S104: when i>2, according to A i-1 and A i-2 The second arrival time is calculated based on the gas flow rate obtained each time, the recorded acquisition time and the first distance interval L1, and the second arrival time and A are calculated based on the second arrival time and A. i-1 The acquisition time of the record is in A i-1 After each acquisition, the gas pressure, gas flow rate and gas temperature are acquired after a delay of the second arrival time, and the acquisition time and the number of acquisitions are recorded;

[0086] Step S105, marking the gas pressures, corresponding acquisition times, and acquisition times acquired by all measuring sensors as first gas pressure data;

[0087] Step S106, marking the ambient temperature acquired by the temperature sensors corresponding to all the measuring sensors and the corresponding acquisition time and acquisition number as first ambient temperature data;

[0088] Step S107, record the occurrence of gas leakage in the gas pipeline as M=1, record the occurrence of gas leakage in the gas pipeline as M=0, and mark whether the gas pipeline leaks when the first gas pressure data and the first pressure and temperature data are obtained; and merge and store according to the corresponding measurement sensors, and record them as the first pressure and temperature data.

[0089] Step S2, performing pressure error correction processing on the first gas pressure data based on the first ambient temperature data, and calculating the gas leakage risk index to obtain first gas leakage characteristic data; Step S2 includes the following sub-steps:

[0090] Step S201, setting the standard gas temperature to K0; based on the first ambient temperature data and the first gas pressure data, and according to the gas pressure and gas temperature at the same acquisition time, using the pressure correction formula to calculate the gas pressure at the standard gas temperature, recorded as the standard gas pressure, the pressure correction formula is as follows: Where Pe represents the gas pressure at the standard gas temperature, Pa represents the obtained gas pressure, and Ka represents the gas temperature corresponding to the obtained gas pressure;

[0091] Step S202, replacing the corresponding gas pressure with the calculated standard gas pressure, and obtaining the second gas pressure data after completion;

[0092] Step S203, for any two adjacent positions on the gas pipeline, set as i and i+1, calculate the gas leakage risk index according to the corresponding second gas pressure data, and obtain the first gas leakage characteristic data after completion; the gas leakage risk index calculation includes calculating the pressure difference characteristic, calculating the pressure difference change rate characteristic, calculating the pressure difference gradient characteristic and calculating the pressure difference fluctuation characteristic; Step S203 includes the following sub-steps:

[0093] Step S2031, calculating the pressure difference feature includes: assuming that the second gas pressure data corresponding to the pressure sensors at any two adjacent positions on the gas pipeline are P i，j and P i+1，j , where j represents the number of acquisitions; based on P i，j and P i+1，j The pressure difference characteristic is calculated using the pressure difference calculation formula. The pressure difference calculation formula is as follows: B i,j =|P i,j -P i+1,j |, where B i，j Represents the pressure difference characteristics between the i-th position and the i+1-th position on the gas pipeline at the j-th acquisition;

[0094] Step S2032, calculating the pressure difference change rate characteristic includes: assuming P i，j The corresponding acquisition time is t i，j, P i，j-1 The acquisition time is t i，j-1 , the pressure difference change rate characteristic is calculated using the pressure difference change rate calculation formula. The pressure difference change rate calculation formula is as follows: Among them C i，j Representative B i，j With B i，j+1 The pressure difference change rate characteristic, B i，j+1 Represents the pressure difference characteristics between the i-th position and the i+1-th position on the gas pipeline at the j-1th acquisition;

[0095] Step S2033, calculating the pressure difference gradient feature includes: based on the pressure difference feature B i，j And the first distance interval L1, the pressure difference gradient characteristic is calculated using the pressure difference gradient calculation formula, and the pressure difference gradient calculation formula is as follows: Where D i，j Representative B i，j Pressure gradient characteristics;

[0096] Step S2034, calculating the pressure difference fluctuation characteristics includes: setting the first time length as T0, obtaining the pressure difference characteristics within the first time length, and sorting them according to the corresponding time series, marking them as the first pressure difference characteristic sequence, and calculating the pressure difference fluctuation characteristics within the first time length using the pressure difference fluctuation calculation formula, the pressure difference fluctuation calculation formula is as follows: Where E i，j represents the pressure difference fluctuation feature of the i-th position within the corresponding first time length with the j-th acquisition corresponding time as the end point, Bn represents the n-th pressure difference feature in the first pressure difference feature sequence, B0 represents the average value of the first pressure difference feature sequence; N represents the total number of pressure difference features in the first pressure difference feature sequence;

[0097] Step S204: marking the pressure difference feature, the pressure difference change rate feature, the pressure difference gradient feature, and the pressure difference fluctuation feature as gas leakage risk indicators.

[0098] Step S3, constructing a gas leakage risk quantification model, and calculating the gas leakage risk based on the gas leakage risk quantification model to obtain gas leakage risk data; Step S3 includes the following sub-steps:

[0099] Step S301, constructing a leakage risk logistic regression model based on the logistic regression model. The leakage risk logistic regression model is as follows: Where Y(M=1) represents the risk probability of leakage; B, C, D and E represent the pressure difference characteristic, the pressure difference change rate characteristic, the pressure difference gradient characteristic and the pressure difference fluctuation characteristic in order, respectively; β0, β1, β2, β3 and β4 are the parameters of the leakage risk logistic regression model;

[0100] Step S302, setting the learning rate of model training to h1, the training round to h2, and the batch size to h3, training the leakage risk logistic regression model using the first gas leakage feature data, and obtaining a gas leakage risk quantification model;

[0101] Step S303, input the most recently acquired first gas leakage characteristic data into the gas leakage risk quantification model at a second time interval to obtain the gas leakage risks of all corresponding positions on the gas pipeline, marked as gas leakage risk data, the second time interval is t2, t2>t1, and the gas leakage risk range is [0, 1].

[0102] Step S4, based on the gas leakage risk data, a gas leakage risk warning is performed on the gas pipeline; Step S4 includes the following sub-steps:

[0103] Step S401, setting the risk threshold to Y0, and judging whether any gas leakage risk satisfies or is greater than or equal to Y0 based on the gas leakage risk data;

[0104] Step S402: If the value is greater than or equal to Y0, an alarm signal is triggered, and the position on the gas pipeline corresponding to the gas leakage risk greater than or equal to Y0 is marked as a dangerous position, and the position on the gas pipeline next to the dangerous position is also marked as a dangerous position, and the gas pipeline between the two dangerous positions is marked as a leaking pipeline;

[0105] Step S403: If it does not satisfy the value greater than or equal to Y0, for the gas leakage risk corresponding to any position on the gas pipeline, determine whether it satisfies the value greater than or equal to 0.85*Y0 and increases for N1 consecutive times.

[0106] Step S404: If the conditions are met, a warning signal is triggered, and the corresponding position on the gas pipeline is marked as a risk position, the position on the gas pipeline next to the risk position is also marked as a risk position, and the gas pipeline between the two risk positions is marked as a risk pipeline;

[0107] Step S405: if not satisfied, the corresponding position on the gas pipeline is marked as risk-free.

[0108] Example 3, please refer to Figure 5 As shown, Figure 5The structural schematic diagram of an electronic device is illustrated, and the electronic device may include: a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus. The memory stores computer-readable instructions, and the processor may call the instructions in the memory. When the computer-readable instructions are executed by the processor, the steps in the dynamic risk early warning method for gas pipelines based on the Internet of Things are executed to achieve the following functions: obtaining the first pressure and temperature data of the gas pipeline; the first pressure and temperature data include the first gas pressure data and the first ambient temperature data; performing pressure error correction processing on the first gas pressure data based on the first ambient temperature data, and calculating the gas leakage risk index to obtain the first gas leakage characteristic data; constructing a gas leakage risk quantification model, and calculating the gas leakage risk based on the gas leakage risk quantification model to obtain the gas leakage risk data; and performing a gas leakage risk early warning on the gas pipeline based on the gas leakage risk data.

[0109] In addition, the logic instructions in the above-mentioned memory can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art, and the computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk.

[0110] Embodiment 4, the present application also provides a computer-readable storage medium, the present application provides a storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned dynamic risk warning method for gas pipelines based on the Internet of Things are executed to achieve the following functions: obtaining first pressure and temperature data of the gas pipeline; the first pressure and temperature data include first gas pressure data and first ambient temperature data; performing pressure error correction processing on the first gas pressure data based on the first ambient temperature data, and calculating a gas leakage risk index to obtain first gas leakage characteristic data; constructing a gas leakage risk quantification model, and calculating the gas leakage risk based on the gas leakage risk quantification model to obtain gas leakage risk data; and performing a gas leakage risk warning on the gas pipeline based on the gas leakage risk data.

[0111] Through the description of the above implementation methods, the embodiments of the present invention can be provided as methods, systems or computer program products. Based on such an understanding, the above technical solutions can be essentially or partly contributed to the prior art in the form of software products, which can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and include several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0112] In the embodiments provided in the present application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of systems, modules and units can be electrical, mechanical or other forms.

[0113] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A gas pipeline dynamic risk early warning platform based on the Internet of Things, characterized by: It includes data acquisition module, data processing module, risk quantification module and dynamic early warning module; The data acquisition module is used to acquire first pressure and temperature data of the gas pipeline; the first pressure and temperature data includes first gas pressure data and first ambient temperature data; The data processing module includes an error correction unit and a feature calculation unit. The error correction unit performs pressure error correction processing on the first gas pressure data based on the first ambient temperature data. The feature calculation unit is used to calculate the gas leakage risk index to obtain the first gas leakage feature data. The risk quantification module includes a model building unit and a risk quantification unit. The model building unit is used to build a gas leakage risk quantification model. The risk quantification unit calculates the gas leakage risk based on the gas leakage risk quantification model to obtain gas leakage risk data. The dynamic early warning module provides a gas leakage risk early warning for the gas pipeline based on the gas leakage risk data.

2. According to the Internet of Things-based gas pipeline dynamic risk early warning platform of claim 1, it is characterized in that: The data acquisition module is configured with a data acquisition strategy, which includes: A pressure sensor, a flow rate sensor and a temperature sensor are arranged at a first distance interval based on the gas pipeline. The pressure sensor, the flow rate sensor and the temperature sensor are arranged at the same position, and the first distance interval is L1; the pressure sensor, the flow rate sensor and the temperature sensor are marked as measurement sensors, and according to the order of the arrangement positions of the measurement sensors, the measurement sensor at any position is marked as A i , where i represents the i-th position, 1≤i.

3. According to the Internet of Things-based gas pipeline dynamic risk early warning platform of claim 2, it is characterized in that: The data acquisition module is configured with a data acquisition strategy, which also includes: For measuring sensor A i , when i=1, the gas pressure, gas flow rate and gas temperature in the gas pipeline are obtained at a first time interval, and the acquisition time and the number of acquisitions are recorded. The first time interval is t1; When i=2, the first arrival time is calculated according to the flow rate of the gas obtained by A2 each time and the first distance interval L1, and according to the first arrival time and the acquisition time recorded by A1, the gas pressure, gas flow rate and gas temperature in the gas pipeline are acquired after the first arrival time length is delayed after each acquisition by A1, and the acquisition time and the number of acquisitions are recorded; When i>2, according to A i-1 and A i-2 The second arrival time is calculated based on the gas flow rate obtained each time, the recorded acquisition time and the first distance interval L1, and the second arrival time and A are calculated based on the second arrival time and A. i-1 The acquisition time of the record is in A i-1 After each acquisition, the gas pressure, gas flow rate and gas temperature are acquired after a delay of the second arrival time, and the acquisition time and the number of acquisitions are recorded; The gas pressures, corresponding acquisition times, and acquisition times acquired by all measuring sensors are marked as first gas pressure data; Marking the ambient temperature acquired by the temperature sensors corresponding to all the measuring sensors and the corresponding acquisition time and acquisition number as first ambient temperature data; Gas leakage in the gas pipeline is recorded as M=1, and no gas leakage in the gas pipeline is recorded as M=0. The gas pipeline is marked according to whether leakage occurs when the first gas pressure data and the first pressure and temperature data are obtained; and the corresponding measurement sensors are combined and stored and recorded as the first pressure and temperature data.

4. According to the Internet of Things-based gas pipeline dynamic risk early warning platform of claim 3, it is characterized in that: The error correction unit is configured with an error correction strategy, which includes: The standard gas temperature is set to K0; based on the first ambient temperature data and the first gas pressure data, and according to the gas pressure and gas temperature at the same acquisition time, the gas pressure at the standard gas temperature is calculated using the pressure correction formula, which is recorded as the standard gas pressure. The pressure correction formula is as follows: Wherein Pe represents the gas pressure at the standard gas temperature, Pa represents the obtained gas pressure, and Ka represents the gas temperature corresponding to the obtained gas pressure; and the corresponding gas pressure is replaced with the calculated standard gas pressure, and the second gas pressure data is obtained after completion.

5. The gas pipeline dynamic risk early warning platform based on the Internet of Things according to claim 4 is characterized in that: The feature calculation unit is configured with a feature calculation strategy, which includes: For any two adjacent positions on the gas pipeline, set as i and i+1, the gas leakage risk index is calculated according to the corresponding second gas pressure data, and the first gas leakage characteristic data is obtained after completion; the gas leakage risk index calculation includes calculating the pressure difference characteristic, calculating the pressure difference change rate characteristic, calculating the pressure difference gradient characteristic and calculating the pressure difference fluctuation characteristic; Calculating the pressure difference feature includes: assuming that the second gas pressure data corresponding to the pressure sensors at any two adjacent positions on the gas pipeline are P i，j and P i+1，j , where j represents the number of acquisitions; based on P i，j and P i+1，j The pressure difference characteristic is calculated using the pressure difference calculation formula. The pressure difference calculation formula is as follows: B i,j =P i,j -P i+ 1 ,j |, where B i，j Represents the pressure difference characteristics between the i-th position and the i+1-th position on the gas pipeline at the j-th acquisition; Calculating the pressure difference change rate characteristics includes: assuming P i，j The corresponding acquisition time is t i，j , P i，j-1 The acquisition time is t i，j-1 , the pressure difference change rate characteristic is calculated using the pressure difference change rate calculation formula. The pressure difference change rate calculation formula is as follows: Among them C i，j Representative B i，j With B i，j+1 The pressure difference change rate characteristic, B i，j+1 Represents the pressure difference characteristics between the i-th position and the i+1-th position on the gas pipeline during the j-1-th acquisition.

6. The gas pipeline dynamic risk early warning platform based on the Internet of Things according to claim 5 is characterized in that: The feature calculation unit is configured with a feature calculation strategy, which also includes: Calculating the pressure difference gradient feature includes: based on the pressure difference feature B i，j And the first distance interval L1, the pressure difference gradient characteristic is calculated using the pressure difference gradient calculation formula, and the pressure difference gradient calculation formula is as follows: Where D i，j Representative B i，j Pressure gradient characteristics; Calculating the pressure difference fluctuation characteristics includes: setting the first time length as T0, obtaining the pressure difference characteristics within the first time length, and sorting them according to the corresponding time series, marking them as the first pressure difference characteristic sequence, and calculating the pressure difference fluctuation characteristics within the first time length using the pressure difference fluctuation calculation formula. The pressure difference fluctuation calculation formula is as follows: Where E i，j represents the pressure difference fluctuation feature of the i-th position within the corresponding first time length with the j-th acquisition corresponding time as the end point, Bn represents the n-th pressure difference feature in the first pressure difference feature sequence, B0 represents the average value of the first pressure difference feature sequence; N represents the total number of pressure difference features in the first pressure difference feature sequence; The pressure difference characteristics, pressure difference change rate characteristics, pressure difference gradient characteristics and pressure difference fluctuation characteristics are marked as gas leakage risk indicators.

7. The gas pipeline dynamic risk early warning platform based on the Internet of Things according to claim 6 is characterized in that: The model building unit is configured with a model building strategy, which includes: A leakage risk logistic regression model is constructed based on the logistic regression model. The leakage risk logistic regression model is as follows: Where Y(M=1) represents the risk probability of leakage; B, C, D and E represent the pressure difference characteristic, the pressure difference change rate characteristic, the pressure difference gradient characteristic and the pressure difference fluctuation characteristic in order, respectively; β0, β1, β2, β3 and β4 are the parameters of the leakage risk logistic regression model; The learning rate of model training is set to h1, the training round is set to h2, and the batch size is set to h3. The leakage risk logistic regression model is trained using the first gas leakage feature data to obtain a gas leakage risk quantification model.

8. The gas pipeline dynamic risk early warning platform based on the Internet of Things according to claim 7 is characterized in that: The risk quantification unit is equipped with risk quantification strategies, which include: The most recently acquired first gas leakage characteristic data is input into the gas leakage risk quantification model at a second time interval to obtain the gas leakage risk of all corresponding positions on the gas pipeline, which is marked as gas leakage risk data. The second time interval is t2, t2>t1, and the gas leakage risk range is [0, 1].

9. The gas pipeline dynamic risk early warning platform based on the Internet of Things according to claim 8 is characterized in that: The dynamic warning module is configured with dynamic warning strategies, which include: Set the risk threshold to Y0, and based on the gas leakage risk data, determine whether any gas leakage risk satisfies or exceeds Y0; if it satisfies or exceeds Y0, trigger an alarm signal, and mark the position on the gas pipeline corresponding to the gas leakage risk greater than or equal to Y0 as a dangerous position, and also mark the position on the gas pipeline next to the dangerous position as a dangerous position, and mark the gas pipeline between the two dangerous positions as a leaking pipeline; If it does not meet the requirement of being greater than or equal to Y0, for the gas leakage risk corresponding to any position on the gas pipeline, determine whether it meets the requirement of being greater than or equal to 0.85*Y0 and increases for N1 consecutive times. If it meets the requirement, the early warning signal is triggered, and the corresponding position on the gas pipeline is marked as a risk position, and the position after the risk position on the gas pipeline is also marked as a risk position, and the gas pipeline between the two risk positions is marked as a risk pipeline; if it does not meet the requirement, the corresponding position on the gas pipeline is marked as non-risk.

10. A gas pipeline dynamic risk early warning method based on the Internet of Things, applicable to the gas pipeline dynamic risk early warning platform based on the Internet of Things according to any one of claims 1 to 9, characterized in that: The steps include: Acquire first pressure and temperature data of the gas pipeline; the first pressure and temperature data includes first gas pressure data and first ambient temperature data; Performing pressure error correction processing on the first gas pressure data based on the first ambient temperature data, and calculating the gas leakage risk index to obtain first gas leakage characteristic data; Construct a gas leakage risk quantification model, and calculate the gas leakage risk based on the gas leakage risk quantification model to obtain gas leakage risk data; Provide gas leakage risk warning for gas pipelines based on gas leakage risk data.

Images