A method for processing liquefied natural gas safety monitoring data

Abstract

The present invention relates to the technical field of data monitoring, and particularly relates to a method for processing liquefied natural gas safety monitoring data, including: determining a suspected interference period and a normal period according to an initial interference factor, obtaining the maximum interference of a merged segment, and further obtaining the first data interference degree of the merged segment; obtaining a combustion sufficiency factor sequence and an initial interference factor sequence corresponding to a suspected interference period with a duration greater than a preset duration threshold to obtain a second data interference degree, and obtaining a final relative interference degree according to a standard combustion sufficiency factor and the second data interference degree; obtaining an interference filtering factor according to the final relative interference degree and the first data interference degree, further obtaining the step size of a filter, and using the filter to filter the liquefied natural gas flow data at each moment and perform anomaly detection on the liquefied natural gas flow data. The present invention can achieve precise filtering, making the anomaly detection of liquefied natural gas flow data more accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of data monitoring, and particularly to a method for processing liquefied natural gas safety monitoring data. Background Art

[0002] Liquefied Natural Gas (LNG), as a clean energy source, has demonstrated remarkable advantages in the power generation field. As a power generation fuel, LNG has been widely used in the power industry due to its high combustion efficiency and low pollution emissions. However, given the low temperature and flammable nature of LNG, its safety monitoring work is particularly important.

[0003] In an LNG power station, accurately monitoring the LNG flow rate from the storage tank to the vaporizer is crucial for ensuring the stable operation of the power station and continuously outputting stable electricity. This monitoring process not only helps maintain the correct LNG supply rate required by the vaporizer but also ensures the ideal ratio of air to fuel, which is essential for achieving efficient and environmentally friendly combustion. However, during the flow of LNG, problems such as eddies and vortices may occur, which can directly affect the accuracy of flow measurement. In addition, the violent reaction generated by the mixture of LNG and air after vaporization may cause vibrations in the LNG input pipeline, thereby interfering with the acquisition of flow data. These interfering data not only reduce the accuracy of flow monitoring but may also have an adverse impact on the safety monitoring of liquefied natural gas. Therefore, in an LNG power station, effective measures must be taken to ensure the accurate monitoring of the LNG flow rate, thereby ensuring the safe and stable operation of the power station. Summary of the Invention

[0004] In order to solve the technical problem of inaccurate liquefied natural gas monitoring, the purpose of the present invention is to provide a method for processing liquefied natural gas safety monitoring data, and the specific technical solutions adopted are as follows:

[0005] Determine the initial interference factor corresponding to each liquefied natural gas flow data according to the liquefied natural gas flow data and the power generation data, determine the suspected interference period and the normal period, and then obtain the merged segment, the maximum interference of the merged segment, and the first data interference degree;

[0006] Construct a combustion sufficiency factor sequence and an initial interference factor sequence respectively for the liquefied natural gas flow data corresponding to each moment in the suspected interference period with a duration greater than the preset duration threshold, and then obtain the second data interference degree. Obtain the final relative interference degree according to the mean value of the combustion sufficiency factor corresponding to the liquefied natural gas flow data corresponding to each moment in the normal period and the second data interference degree;

[0007] An interference filtering factor is obtained based on the final relative interference degree, duration, and the first data interference degrees, durations, and the total duration of the power generation process of each of them, and then the step size of the filter is obtained. After filtering the liquefied natural gas flow rate data at each moment using the filter, anomaly detection is performed to obtain anomaly values, and the liquefied natural gas flow rate data corresponding to the anomaly values greater than the preset anomaly threshold is abnormal data.

[0008] Further, the process of obtaining the initial interference factor includes:

[0009] The power generation data corresponding to the liquefied natural gas flow rate data at each moment is obtained to form a power generation set;

[0010] The liquefied natural gas flow rate data corresponding to the same power generation data in the power generation set is selected to form a liquefied natural gas flow rate data set;

[0011] The mode in the liquefied natural gas flow rate data set is obtained as the standard liquefied natural gas flow rate data;

[0012] The normalized value of the absolute value of the difference between each liquefied natural gas flow rate data in the liquefied natural gas flow rate data set and the standard liquefied natural gas flow rate data is calculated as the initial interference factor.

[0013] Further, the process of obtaining the suspected interference period and the normal period includes:

[0014] The moment when the initial interference factor is greater than the first preset value or less than the second preset value is used as the suspected interference moment;

[0015] Adjacent suspected interference moments constitute the suspected interference period, and adjacent non-suspected interference moments constitute the normal period.

[0016] Further, obtaining the maximum interference of the merged segment includes:

[0017] Obtain the first minimum value and the first maximum value of the initial interference factor corresponding to the merged segment;

[0018] Obtain the second minimum value between the first minimum value and the value obtained by subtracting the first maximum value from the third preset value;

[0019] The value obtained by subtracting the second minimum value from the third preset value is used as the maximum interference of the merged segment.

[0020] Further, obtaining the first data interference degree of the merged segment includes:

[0021] Dividing the number of interference moments within the merged segment by the duration of the merged segment to obtain a duration ratio, where the interference moment is a suspected interference moment within the suspected interference period;

[0022] Obtaining the sum of the time intervals between adjacent suspected interference periods in the merged segment;

[0023] Multiplying the duration ratio, the maximum interference, and the reciprocal of the sum of the time intervals in sequence to obtain the first data interference degree.

[0024] Further, the process of obtaining the combustion sufficiency factor includes:

[0025] Obtaining the first oxygen content data, the first carbon monoxide content data in the air before combustion, the second oxygen content data, and the second carbon monoxide content data in the air after combustion corresponding to the liquefied natural gas flow data at any moment in any suspected interference period;

[0026] After subtracting the second oxygen content data from the first oxygen content data, dividing the result by the first oxygen content data to obtain an oxygen content ratio;

[0027] After subtracting the first carbon monoxide content data from the second carbon monoxide content data, dividing the result by the second carbon monoxide data to obtain a carbon monoxide content ratio;

[0028] The value obtained by subtracting the carbon monoxide content ratio from the third preset value is used as the carbon monoxide content change characteristic value, and the oxygen content ratio is used as the oxygen content change characteristic value;

[0029] Calculating the average value of the carbon monoxide content change characteristic value and the oxygen content change characteristic value as the combustion sufficiency factor.

[0030] Further, the process of obtaining the second data interference degree includes:

[0031] Calculating the Pearson correlation coefficient between the combustion sufficiency factor sequence and the initial interference factor sequence and performing normalization processing to obtain the second data interference degree of the suspected interference periods with a duration greater than a preset duration threshold.

[0032] Further, the process of obtaining the final relative interference degree includes:

[0033] Calculating the second maximum value of the absolute value of the difference between the combustion sufficiency factor corresponding to the liquefied natural gas flow data and the standard combustion sufficiency factor in any suspected interference period;

[0034] Normalize the product of the second maximum value and the duration of the suspected interference period to obtain a normalized value. Multiply the normalized value by the second data interference degree, and then add the second data interference degree to obtain the final relative interference degree.

[0035] Further, the process of obtaining the interference filtering factor includes:

[0036] Divide the total duration of the merged segment by the total duration of the power generation process, and then multiply by the average value of the first data interference degree of the merged segment to obtain the first factor;

[0037] Divide the total duration of the suspected interference periods with a duration greater than the preset duration threshold by the total duration of the power generation process, and then multiply by the average value of the final relative interference degree of the suspected interference periods with a duration greater than the preset duration threshold to obtain the second factor;

[0038] Add the first factor and the second factor to obtain the interference filtering factor.

[0039] Further, obtaining the step size of the filter according to the interference filtering factor includes:

[0040] Normalize the interference filtering factor to obtain a normalized interference filtering value, and multiply the normalized interference filtering value by a preset weight to obtain a step size adjustment value;

[0041] Subtract the step size adjustment value from the preset initial step size to obtain the step size of the filter.

[0042] The present invention has the following beneficial effects:

[0043] First, determine the initial interference factor corresponding to each liquefied natural gas flow data according to the liquefied natural gas flow data and the power generation data, and determine the suspected interference periods and normal periods, and then obtain the merged segment, the maximum interference of the merged segment, and the first data interference degree. The merged segment is the period of normal combustion during the power generation process. The greater the first data interference degree, the more serious the interference effect on the liquefied natural gas flow data corresponding to the merged segment.

[0044] Secondly, for each moment in the suspected interference period with a duration greater than the preset duration threshold, the combustion sufficiency factors corresponding to the liquefied natural gas flow rate data and the initial interference factors respectively form a combustion sufficiency factor sequence and an initial interference factor sequence. Furthermore, a second data interference degree is obtained. The final relative interference degree is obtained based on the mean value of the combustion sufficiency factors corresponding to the liquefied natural gas flow rate data at each moment in the normal period and the second data interference degree. The suspected interference period with a duration greater than the preset duration threshold is the period of abnormal combustion during the power generation process. The greater the final relative interference degree, the more severely the liquefied natural gas flow rate data corresponding to the suspected interference period with a duration greater than the preset duration threshold is affected by interference.

[0045] Finally, based on the final relative interference degree, duration, and each of the first data interference degrees, durations, and the total duration of the power generation process, an interference filtering factor is obtained. Furthermore, the step size of the filter is obtained, and after filtering the liquefied natural gas flow rate data at each moment using the filter, an anomaly detection is performed to obtain anomaly values. The liquefied natural gas flow rate data corresponding to the anomaly values greater than the preset anomaly threshold is abnormal data. The interference filtering factor is used to adjust the step size of the filter for precise filtering. Performing anomaly detection on the liquefied natural gas flow rate data is a prior art.

[0046] In summary, the present invention can accurately adjust the step size of the filter based on the degree of interference suffered by the liquefied natural gas flow rate data to achieve precise filtering, making the anomaly detection of the filtered liquefied natural gas flow rate data more accurate. And if an anomaly is detected, an alarm can be issued in a timely manner and corresponding measures can be taken. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0048] Figure 1 It is a flowchart of a method for processing liquefied natural gas safety monitoring data provided by the first embodiment of the present invention;

[0049] Figure 2 It is a flowchart of the process for obtaining the initial interference factor provided by the second embodiment of the present invention;

[0050] Figure 3 It is a flowchart of determining the suspected interference period and the normal period according to the initial interference factor provided by the third embodiment of the present invention;

[0051] Figure 4 Flow chart of the process for obtaining the maximum interference provided by the fourth embodiment of the present invention;

[0052] Figure 5 Flow chart of the process for obtaining the degree of interference of the first data provided by the fifth embodiment of the present invention;

[0053] Figure 6 Flow chart of the process for obtaining the combustion sufficiency factor provided by the sixth embodiment of the present invention;

[0054] Figure 7 Flow chart of the process for obtaining the final relative interference degree provided by the seventh embodiment of the present invention;

[0055] Figure 8 Flow chart of the process for obtaining the interference filtering factor provided by the eighth embodiment of the present invention;

[0056] Figure 9 Flow chart of the process for obtaining the step size of the filter provided by the ninth embodiment of the present invention. Detailed implementation manners

[0057] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following, in combination with the accompanying drawings and preferred embodiments, details a liquefied natural gas safety monitoring data processing method proposed according to the present invention, including its specific implementation manners, structures, features and effects. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0058] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0059] The following specifically describes the specific solution of a liquefied natural gas safety monitoring data processing method provided by the present invention with reference to the accompanying drawings.

[0060] Please refer to Figure 1 , which shows a flow chart of a liquefied natural gas safety monitoring data processing method provided by the first embodiment of the present invention. The method includes:

[0061] S101. Determine the initial interference factor corresponding to each liquefied natural gas flow data according to the liquefied natural gas flow data and power generation data, determine the suspected interference period and the normal period, and then obtain the merged segment, the maximum interference of the merged segment and the degree of interference of the first data.

[0062] In the power generation process of an LNG power station, the present invention implements a set of precise data acquisition systems to track and record key parameters in real time. The system automatically collects data every 5 seconds, including liquefied natural gas flow data, oxygen content data and carbon monoxide content data before and after the gasified natural gas is mixed with air for combustion corresponding to the liquefied natural gas flow data, and the corresponding power generation data. Specifically, the liquefied natural gas flow data is monitored by a high-precision flowmeter, while the oxygen content data and carbon monoxide content data are measured by an advanced gas analyzer, and the power generation data is directly read by an electric meter.

[0063] To ensure the stability of combustion efficiency, a fixed pre-combustion air-fuel ratio is adopted to ensure the full mixing of fuel and oxygen. This ratio is preset based on the liquefied natural gas flow data at each moment to ensure that the supply of the mixed air is known and appropriate.

[0064] It should be noted that there are certain time delays in the series of processes from the change in liquefied natural gas flow data to the change in oxygen content data and carbon monoxide content data after its gasification and mixing with air for combustion, and then to the generation of final electricity. These delays have been accurately measured through actual tests and calibrations and integrated into the algorithm of the combustion control system. Operators and technicians can use this information to more precisely adjust the fuel supply and air flow, thereby maintaining the ideal combustion efficiency and stable power generation.

[0065] The specific process of power generation is as follows:

[0066] 1. Liquefied natural gas is transported to the power station by tanker trucks and stored in cryogenic storage tanks.

[0067] 2. The stored liquefied natural gas is vaporized through a vaporizer into a gaseous state.

[0068] 3. The vaporized liquefied natural gas is transported to a gas turbine. In the gas turbine, the liquefied natural gas is mixed with air and burned to produce high-temperature and high-pressure gas, which drives the blades of the gas turbine to rotate, thereby driving the generator connected thereto to generate electricity.

[0069] 4. The exhaust gas discharged from the gas turbine still has a relatively high temperature, and this part of the heat can be recovered by a waste heat boiler to generate steam.

[0070] 5. The generated steam is then sent to a steam turbine, and the steam drives the blades of the steam turbine to rotate, further driving the generator to generate electricity.

[0071] Under normal conditions, there is a direct correlation between the liquefied natural gas flow data and the power generation data. Specifically, an increase in the liquefied natural gas flow often means that the combustion system of the power station receives more fuel supply, thereby prompting a corresponding increase in power generation. Conversely, when the liquefied natural gas flow decreases, the fuel supply decreases, resulting in a reduction in power generation.

[0072] In chronological order, starting from the first suspected interference period with a duration less than the preset duration threshold, merge until a suspected interference period with a duration not less than the preset duration threshold is encountered (the merge does not include the suspected interference period with a duration not less than the preset duration threshold), stop the merge, obtain a time - continuous merged segment, and then, in the subsequent time, continue to repeat this merge operation to obtain several non - overlapping merged segments in time. The merged segments represent the fully - burned (normal) periods.

[0073] The process of obtaining the initial interference factor will be described in detail in the second embodiment and will not be elaborated here.

[0074] Determining the suspected interference periods and normal periods based on the initial interference factor will be described in detail in the third embodiment and will not be elaborated here.

[0075] The process of obtaining the maximum interference will be described in detail in the fourth embodiment and will not be elaborated here.

[0076] The process of obtaining the first - degree data interference will be described in detail in the fifth embodiment and will not be elaborated here.

[0077] S102. For the suspected interference periods with a duration greater than the preset duration threshold, respectively form a fully - burned factor sequence and an initial interference factor sequence for the liquefied natural gas flow data at each moment, and then obtain the second - degree data interference. Based on the mean value of the fully - burned factors corresponding to the liquefied natural gas flow data at each moment in the normal periods and the second - degree data interference, obtain the final relative interference degree.

[0078] The process of obtaining the fully - burned factor will be described in detail in the sixth embodiment and will not be elaborated here.

[0079] Specifically, the process of obtaining the second - degree data interference includes:

[0080] Calculate the Pearson correlation coefficient between the fully - burned factor sequence and the initial interference factor sequence and perform normalization processing to be used as the second - degree data interference for the suspected interference periods with a duration greater than the preset duration threshold.

[0081] Among them, the Pearson correlation coefficient is a well-known technology, and its value range is between -1 and 1. The closer it is to -1, the more negatively correlated the two sequences are.

[0082] The second data interference degree can be expressed as:

[0083] ;

[0084] Among them, the represents the Pearson correlation coefficient corresponding to the x-th suspected interference period with a duration greater than the preset duration threshold, and the represents the second data interference degree corresponding to the x-th suspected interference period with a duration greater than the preset duration threshold, and the represents a normalization function, such as a range normalization function.

[0085] Since in the case of no interference and only affected by combustion, the initial interference factor should be negatively correlated with the combustion sufficiency factor, so the larger it is, the more serious the interference in the x-th suspected interference period. When the second data interference degree is larger, more precise interference filtering is required to reduce errors.

[0086] The process of obtaining the final relative interference degree will be described in detail in the seventh embodiment and will not be elaborated here.

[0087] S103. Obtain an interference filtering factor according to the final relative interference degree, duration, and each of the first data interference degrees, durations, and the total duration of the power generation process, and then obtain the step size of the filter. After filtering the liquefied natural gas flow data at each moment using the filter, perform anomaly detection to obtain anomaly values, and the liquefied natural gas flow data corresponding to the anomaly values greater than the preset anomaly threshold is abnormal data.

[0088] The process of obtaining the interference filtering factor will be described in detail in the eighth embodiment and will not be elaborated here.

[0089] The process of obtaining the step size will be described in detail in the ninth embodiment and will not be elaborated here.

[0090] Anomaly detection uses a data anomaly detection algorithm, and the data anomaly detection algorithm is a prior art, such as the LOF algorithm, and algorithms such as Z-Score, box plot, and KNN can also be used.

[0091] When the data anomaly detection algorithm is the LOF algorithm, the anomaly threshold can be set independently, and preferably it is 2.

[0092] When abnormal liquefied natural gas flow data appears, an alarm can be issued in a timely manner and corresponding treatment measures can be taken, such as promptly notifying maintenance personnel to adjust the liquefied natural gas flow and inspecting relevant equipment, etc.

[0093] Figure 2 FIG. is a flowchart of the acquisition process of the initial interference factor provided by the second embodiment of the present invention. The acquisition process of the initial interference factor includes:

[0094] S201. Obtain the power generation data corresponding to the liquefied natural gas flow data at each moment to form a power generation data set.

[0095] S202. Select the liquefied natural gas flow data corresponding to the same power generation data in the power generation data set to form a liquefied natural gas flow data set.

[0096] It should be noted that there can be multiple liquefied natural gas flow data sets. For example, the same power generation data A can form a liquefied natural gas flow data set, and the same power generation data B can also form a liquefied natural gas flow data set.

[0097] S203. Obtain the mode in the liquefied natural gas flow data set as the standard liquefied natural gas flow data.

[0098] It should be noted that if there are multiple modes, obtain the average value of the modes as the standard liquefied natural gas flow data.

[0099] S204. Calculate the normalized value of the absolute value of the difference between each liquefied natural gas flow data in the liquefied natural gas flow data set and the standard liquefied natural gas flow data as the initial interference factor.

[0100] The initial interference factor can be expressed as:

[0101] ;

[0102] where, the represents the b-th liquefied natural gas flow data in the liquefied natural gas flow data set, the is the standard liquefied natural gas flow data, and the represents a normalization function, such as a range normalization function.

[0103] The larger the absolute value of the difference between the liquefied natural gas flow data and the standard liquefied natural gas flow data, the more likely the liquefied natural gas flow data is interference data.

[0104] When liquefied natural gas flows steadily from the storage tank to the vaporizer, the change in power generation efficiency is mainly dominated by combustion efficiency. High-efficiency combustion can ensure that the same amount of liquefied natural gas generates more electricity; conversely, low combustion efficiency will lead to a decrease in power generation efficiency even when the liquefied natural gas flow rate is constant.

[0105] In the operation of a liquefied natural gas power plant, it is crucial to maintain the combustion efficiency at a high and relatively stable level. However, in actual operation, various interference factors may be encountered, such as eddies, vortices, or other flow instabilities generated during the flow of liquefied natural gas, and the violent reactions that may occur when the vaporized liquefied natural gas mixes with air for combustion. These may all cause vibrations in the liquefied natural gas input pipeline, thereby resulting in random interference in data collection. In addition, incomplete combustion (for example, an efficiency exceeding 0.7, which means fuel waste) or abnormally efficient combustion (for example, an efficiency below 0.3, which may cause overload problems, and is more difficult to operate and control precisely, while increasing the risk of equipment damage and safety accidents) are also actual manifestations of combustion instability, and the data of these problems often reflect the actual problems in the combustion process.

[0106] Figure 3 The flowchart for determining the suspected interference period and the normal period according to the initial interference factor provided in the third embodiment of the present invention. The process of obtaining the suspected interference period and the normal period includes:

[0107] S301. When the initial interference factor is greater than the first preset value or less than the second preset value, the corresponding moment is used as the suspected interference moment.

[0108] The first preset value can be set independently, preferably 0.7, and the second preset value can be set independently, preferably 0.3.

[0109] S302. Adjacent suspected interference moments constitute the suspected interference period, and adjacent non-suspected interference moments constitute the normal period.

[0110] Figure 4 The flowchart for the process of obtaining the maximum interference provided in the fourth embodiment of the present invention. Obtaining the maximum interference of the merged segment includes:

[0111] S401. Obtain the first minimum value and the first maximum value of the initial interference factor corresponding to the merged segment.

[0112] is the first minimum value of the initial interference factor in the jth merged segment, is the first maximum value of the initial interference factor in the jth merged segment.

[0113] S402. Obtain the second minimum value between the first minimum value and the value obtained by subtracting the first maximum value from the third preset value.

[0114] The second minimum value can be expressed as:

[0115] ;

[0116] wherein, the represents the function for obtaining the minimum value, and the represents the third preset value, and the third preset value is 1.

[0117] S403. Use the value obtained by subtracting the second minimum value from the third preset value as the maximum interference of the merged segment.

[0118] The maximum interference can be expressed as:

[0119] ;

[0120] wherein, the represents the maximum interference of the j-th merged segment.

[0121] When is closer to 0, it indicates that there is relatively serious incomplete combustion in the j-th merged segment. When is closer to , that is, is closer to 0, it indicates that there is relatively serious abnormal sufficient combustion in the j-th merged segment. Therefore, the smaller , the larger , and the more serious the interference.

[0122] Figure 5 FIG. is a flowchart of the process for obtaining the first data interference degree provided by the fifth embodiment of the present invention. Obtaining the first data interference degree of the merged segment includes:

[0123] S501. Divide the number of interference moments in the merged segment by the duration of the merged segment to obtain a duration ratio, where the interference moment is the suspected interference moment within the suspected interference period.

[0124] The duration ratio can be expressed as: , where the represents the duration of the j-th merged segment, and the represents the number of interference moments in the j-th merged segment, that is, the sum of the durations of the suspected interference periods in the merged segment.

[0125] S502. Obtain the sum of the time intervals between adjacent suspected interference periods in the merged segment.

[0126] The sum of the time intervals can be expressed as:

[0127] ;

[0128] wherein, the is the time interval between the t-th and the (t + 1)-th suspected interference time periods with durations less than the preset duration threshold within the j-th merged segment, and the T represents the number of suspected interference time periods with durations less than the preset duration threshold within the j-th merged segment.

[0129] S503. Multiply the duration ratio, the maximum interference, and the reciprocal of the sum of the time intervals in sequence to obtain the first data interference degree.

[0130] The first data interference degree can be expressed as:

[0131] ;

[0132] wherein, the represents the first data interference degree of the j-th merged segment, the represents the maximum interference of the j-th merged segment, and the represents a normalization function, such as a range normalization function.

[0133] wherein, the The larger it is, the more the proportion of interference moments exists within the j-th merged segment (fully burned (normal) time period), and the more serious the interference is. The smaller it is, that is, The larger it is, the more concentrated the interference moments are within the j-th merged segment, and the more serious the interference effect is.

[0134] Figure 6 FIG.

[0135] S601. Obtain the first oxygen content data, the first carbon monoxide content data in the air before combustion, the second oxygen content data and the second carbon monoxide content data in the air after combustion corresponding to the liquefied natural gas flow data at any moment in any of the suspected interference time periods.

[0136] The first oxygen content data in the air before combustion corresponding to the liquefied natural gas flow data at the y-th moment in the x-th suspected interference time period is expressed as The first carbon monoxide content data is expressed as The second oxygen content data in the air after combustion is expressed as The second carbon monoxide content data is expressed as .

[0137] After subtracting the second oxygen content data from the first oxygen content data, divide the result by the first oxygen content data to obtain an oxygen content ratio.

[0138] The oxygen content ratio can be expressed as:

[0139] ;

[0140] Ideally, complete combustion would consume all the oxygen in the air. That is, if the first oxygen content data before combustion is 20%, the second oxygen content data after complete combustion should be 0%. Therefore, is is 1, indicating complete combustion. If the second oxygen content data remains 5% after combustion, then (15 / 20) is smaller, indicating incomplete combustion.

[0141] After subtracting the first carbon monoxide content data from the second carbon monoxide content data, divide the result by the second carbon monoxide data to obtain a carbon monoxide content ratio.

[0142] The carbon monoxide content ratio can be expressed as:

[0143] ;

[0144] Subtract the carbon monoxide content ratio from the third preset value to obtain a carbon monoxide content change characteristic value, and use the oxygen content ratio as the oxygen content change characteristic value.

[0145] The carbon monoxide content change characteristic value can be expressed as:

[0146] ;

[0147] Among them, the represents the third preset value, and the third preset value is 1.

[0148] Ideally, complete combustion produces carbon dioxide and water vapor, with almost no carbon monoxide or other incompletely burned products. Therefore, if the first carbon monoxide content data before combustion is 5%, and the second carbon monoxide content data remains 5% after complete combustion, then is 0, is , indicating complete combustion. If the second carbon monoxide content data increases to 7% after combustion, it indicates incomplete combustion. At this time The larger (2 / 7) is, and the smaller it is, the less completely the combustion occurs.

[0149] S605. Obtain the average value of the change characteristic value of the carbon monoxide content and the change characteristic value of the oxygen content as the combustion sufficiency factor.

[0150] The combustion sufficiency factor can be expressed as:

[0151] ;

[0152] wherein, the represents the combustion sufficiency factor corresponding to the liquefied natural gas flow rate data at the y-th moment in the x-th suspected interference period.

[0153] Ideally, complete combustion means that all the fuel reacts with sufficient oxygen to produce carbon dioxide and water vapor, with almost no carbon monoxide or other unburned products, that is, the oxygen in the air is consumed completely and the carbon monoxide does not increase. The closer it is to 1, it indicates that the oxygen is consumed completely after combustion, being close to 0 indicates that the carbon monoxide does not increase after combustion. Therefore, the larger it is, the more complete the combustion is. In actual situations, the fuel cannot be 100% burned and will stabilize at a relatively high percentage.

[0154] During the whole process of liquefied natural gas power generation, the power station will strive to maintain the combustion efficiency within an optimal range through precise fuel and air ratio control, efficient combustion equipment and advanced monitoring technologies.

[0155] Therefore, in the case of incomplete or abnormally complete combustion (the suspected interference period with a duration greater than or equal to the preset duration threshold), it is necessary to conduct in-depth analysis to determine the root cause of the abnormal combustion. When filtering out these interference factors, it is necessary to minimize errors as much as possible. For example, when vortices, swirls or other instabilities generated during the flow of liquefied natural gas, as well as the vibration of the input pipeline, may cause interference during both normal and abnormal combustion periods, for the periods of incomplete or abnormally complete combustion, the filtering of interference needs to be particularly precise to ensure the accuracy of the data. For this reason, for the suspected interference period with a duration greater than or equal to the preset duration threshold, more detailed further analysis should be carried out.

[0156] Figure 7 It is a flowchart of the process for obtaining the final relative interference degree provided by the seventh embodiment of the present invention. The process for obtaining the final relative interference degree includes:

[0157] S701. Calculate the second maximum value of the absolute value of the difference between the combustion sufficiency factor corresponding to the liquefied natural gas flow data and the standard combustion sufficiency factor during any of the suspected interference periods.

[0158] The second maximum value can be expressed as:

[0159] ;

[0160] wherein, the represents the second maximum value corresponding to the x-th suspected interference period, the represents the combustion sufficiency factor corresponding to the liquefied natural gas flow data at the y-th moment in the x-th suspected interference period, the represents the standard combustion sufficiency factor, the represents the function for obtaining the maximum value, and the represents the function for obtaining the absolute value.

[0161] The represents the maximum degree of insufficient combustion or the abnormally sufficient combustion degree in the x-th suspected interference period (i.e., a continuous process of insufficient combustion or abnormally sufficient combustion).

[0162] S702. Multiply the second maximum value by the duration of the suspected interference period and then perform normalization processing to obtain a normalized value. Multiply the normalized value by the second data interference degree and then add the second data interference degree to obtain the final relative interference degree.

[0163] The final relative interference degree can be expressed as:

[0164] ;

[0165] wherein, the represents the second data interference degree corresponding to the x-th suspected interference period, the represents the duration of the x-th suspected interference period, the represents the normalization function, such as the range normalization function, and the represents the final relative interference degree corresponding to the x-th suspected interference period.

[0166] Figure 8 is a flowchart of the process for obtaining the interference filtering factor provided in the eighth embodiment of the present invention. The process for obtaining the interference filtering factor includes:

[0167] S801. Divide the total duration of the merged segment by the total duration of the power generation process, and then multiply by the average value of the first data interference degree of the merged segment to obtain a first factor.

[0168] The first factor can be expressed as:

[0169] ;

[0170] wherein, the represents the total duration of the merged segment, the represents the total duration of the power generation process, and the represents the average value of the first data interference degree of the merged segment.

[0171] S802. After dividing the total duration of the suspected interference period with a duration greater than the preset duration threshold by the total duration of the power generation process, multiply by the average value of the final relative interference degree of the suspected interference period with a duration greater than the preset duration threshold to obtain a second factor.

[0172] The second factor can be expressed as:

[0173] ;

[0174] wherein, the represents the total duration of the suspected interference period with a duration greater than the preset duration threshold, the represents the total duration of the power generation process, and the represents the average value of the final relative interference degree of the suspected interference period with a duration greater than the preset duration threshold.

[0175] S803. Add the first factor and the second factor to obtain the interference filtering factor.

[0176] The interference filtering factor can be expressed as:

[0177] ;

[0178] wherein, K represents the interference filtering factor.

[0179] Taking the proportion of the normal combustion duration and abnormal combustion duration in the power generation process and as weights, perform a weighted sum on the interference degree. The larger the K value, the more precise interference filtering is required during the power generation process.

[0180] The filter can be a least mean square error filter to perform interference filtering on the liquefied natural gas flow data during the power generation process, or algorithms such as moving average filtering and exponential smoothing filtering can also be used for interference filtering. The above are all well-known technologies.

[0181] When using the least mean square error filter, the size of the step is the main parameter of the least mean square error filter, and the step is usually selected between 0 and 1. A too small step will result in a slow convergence rate, while a too large step may lead to instability. When there is more data of interest (data when combustion is insufficient or overly sufficient), and the data interference is more severe, a smaller step is required to avoid the filter's overreaction to the interfering data, thereby maintaining the stability of the filter, and a smaller step helps the filter track the desired signal more smoothly. When there is less data of interest and the data interference is not severe, a larger step needs to be selected so that the filter can converge to the optimal solution faster.

[0182] Figure 9 FIG. 4 is a flowchart of the process for obtaining the step size of the filter provided in the ninth embodiment of the present invention. Obtaining the step size of the filter according to the interference filtering factor includes:

[0183] S901. Normalize the interference filtering factor to obtain an interference filtering normalization value, and multiply the interference filtering normalization value by a preset weight to obtain a step size adjustment value.

[0184] The step size adjustment value can be expressed as , where K represents the interference filtering factor, and represents the preset weight, and the preset weight can be set independently, preferably 0.1.

[0185] S902. Subtract the step size adjustment value from the preset initial step size to obtain the step size of the filter.

[0186] The step size can be expressed as:

[0187] ;

[0188] where represents the initial step size, the initial step size can be set independently, preferably 0.2, and M represents the step size.

[0189] When the initial step size is 0.2 and the preset weight is 0.1, the step size can be expressed as:

[0190] .

[0191] The present invention has the following beneficial effects:

[0192] First, determine the initial interference factor corresponding to each liquefied natural gas flow data based on the liquefied natural gas flow data and the power generation data, determine the suspected interference period and the normal period, and then obtain the combined segment, the maximum interference of the combined segment, and the first data interference degree. The combined segment is the period of normal combustion during the power generation process. The greater the first data interference degree, the more serious the interference effect on the liquefied natural gas flow data corresponding to the combined segment.

[0193] Secondly, for the suspected interference periods with a duration greater than the preset duration threshold, respectively form a combustion sufficiency factor sequence and an initial interference factor sequence from the combustion sufficiency factors and the initial interference factors corresponding to the liquefied natural gas flow data at each moment, and then obtain the second data interference degree. Obtain the final relative interference degree based on the mean value of the combustion sufficiency factors corresponding to the liquefied natural gas flow data at each moment in the normal period and the second data interference degree. The suspected interference periods with a duration greater than the preset duration threshold are the periods of abnormal combustion during the power generation process. The greater the final relative interference degree, the more serious the interference effect on the liquefied natural gas flow data corresponding to the suspected interference periods with a duration greater than the preset duration threshold.

[0194] Finally, obtain the interference filtering factor based on the final relative interference degree, duration, and each of the first data interference degrees, durations, and the total duration of the power generation process, and then obtain the step size of the filter. After filtering the liquefied natural gas flow data at each moment using the filter, perform anomaly detection to obtain anomaly values. The liquefied natural gas flow data corresponding to the anomaly values greater than the preset anomaly threshold is abnormal data. The interference filtering factor is used to adjust the step size of the filter for precise filtering. Performing anomaly detection on the liquefied natural gas flow data is a prior art.

[0195] In summary, the present invention can accurately adjust the step size of the filter based on the interference degree of the liquefied natural gas flow data to achieve precise filtering, making the anomaly detection of the filtered liquefied natural gas flow data more accurate. And if an anomaly is detected, an alarm can be issued in a timely manner and corresponding measures can be taken.

[0196] It should be noted that: the above sequence of embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. 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, multitasking and parallel processing are also possible or may be advantageous.

[0197] Each embodiment in this specification is described in a progressive manner. The same or similar parts between each embodiment can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A method for processing liquefied natural gas safety monitoring data, characterized in that: The method comprises: The power generation data corresponding to the liquefied natural gas flow data at each moment is obtained to form a power generation set; the liquefied natural gas flow data corresponding to the same power generation data in the power generation set are selected to form a liquefied natural gas flow data set; the mode in the liquefied natural gas flow data set is obtained as the standard liquefied natural gas flow data; the normalized value of the absolute value of the difference between each liquefied natural gas flow data in the liquefied natural gas flow data set and the standard liquefied natural gas flow data is calculated as the initial interference factor; the moment corresponding to the initial interference factor being greater than a first preset value or less than a second preset value is taken as the suspected interference moment; the adjacent suspected interference moments constitute a suspected interference period, and the adjacent non-suspected interference moments constitute a normal period; in time order, the continuous suspected interference periods whose duration is less than a preset duration threshold are merged to obtain a time-continuous merged segment, and in the subsequent time, the merging operation is continued to be repeated to obtain a number of time-non-repeating merged segments; and then the maximum interference of the merged segment and the first data interference degree are obtained; The combustion sufficiency factor and the initial interference factor corresponding to the liquefied natural gas flow data at each moment in the suspected interference period whose duration is greater than the preset duration threshold are respectively used to form a combustion sufficiency factor sequence and an initial interference factor sequence, thereby obtaining a second data interference degree, and obtaining a final relative interference degree according to the average value of the combustion sufficiency factor corresponding to the liquefied natural gas flow data at each moment in the normal period and the second data interference degree; The interference filtering factor is obtained according to the final relative interference degree, the duration of the suspected interference period and the interference degree of each of the first data, the duration of the merged section, and the total duration of the power generation process, and then the step size of the filter is obtained. The filter is used to filter the liquefied natural gas flow data at each moment and then anomaly detection is performed to obtain anomaly values. The liquefied natural gas flow data corresponding to the anomaly value greater than the preset anomaly threshold is abnormal data.

2. The liquefied natural gas safety monitoring data processing method according to claim 1, characterized in that: Obtaining the maximum interference of the merged segment includes: Obtaining a first minimum value and a first maximum value of the initial interference factor corresponding to the merged segment; Obtain a second minimum value between the first minimum value and a value obtained by subtracting the first maximum value from a third preset value; The value obtained by subtracting the second minimum value from the third preset value is used as the maximum interference of the merged segment.

3. The liquefied natural gas safety monitoring data processing method according to claim 1, characterized in that: Acquiring a first data interference degree of the merged segment includes: The number of interference moments in the merged segment is divided by the duration of the merged segment as a duration ratio, wherein the interference moment is a suspected interference moment in the suspected interference time period; Obtaining the sum of time intervals between adjacent suspected interference time periods in the merged segment; The first data interference degree is obtained by multiplying the duration ratio, the maximum interference, and the inverse of the sum of the time intervals in sequence.

4. The liquefied natural gas safety monitoring data processing method according to claim 1, characterized in that: The process of obtaining the combustion sufficiency factor includes: Obtaining first oxygen content data and first carbon monoxide content data in the air before combustion and second oxygen content data and second carbon monoxide content data in the air after combustion corresponding to the liquefied natural gas flow rate data at any time in any of the suspected interference time periods; The oxygen content ratio is obtained by subtracting the second oxygen content data from the first oxygen content data and then dividing the result by the first oxygen content data; The carbon monoxide content ratio is obtained by subtracting the first carbon monoxide content data from the second carbon monoxide content data and then dividing the subtracted carbon monoxide content data by the second carbon monoxide content data; The value obtained by subtracting the carbon monoxide content ratio from the third preset value is used as the carbon monoxide content change characteristic value, and the oxygen content ratio is used as the oxygen content change characteristic value; The average value of the carbon monoxide content change characteristic value and the oxygen content change characteristic value is calculated as the combustion completeness factor.

5. The liquefied natural gas safety monitoring data processing method according to claim 1, characterized in that: The process of obtaining the second data interference degree includes: The Pearson correlation coefficient between the combustion sufficiency factor sequence and the initial interference factor sequence is calculated and normalized to serve as the second data interference degree of the suspected interference period whose duration is greater than a preset duration threshold.

6. The liquefied natural gas safety monitoring data processing method according to claim 1, characterized in that: The process of obtaining the final relative interference degree includes: Calculating a second maximum value of the absolute value of the difference between the combustion sufficiency factor corresponding to the liquefied natural gas flow rate data and a standard combustion sufficiency factor in any of the suspected interference time periods; The second maximum value is multiplied by the duration of the suspected interference period and then normalized to obtain a normalized value. The normalized value is multiplied by the second data interference degree and then added to obtain the final relative interference degree.

7. The liquefied natural gas safety monitoring data processing method according to claim 1, characterized in that: The process of obtaining the interference filtering factor includes: The first factor is obtained by dividing the total duration of the merging segment by the total duration of the power generation process and then multiplying the result by the average value of the first data interference degree of the merging segment; The second factor is obtained by dividing the total duration of the suspected interference periods whose duration is greater than the preset duration threshold by the total duration of the power generation process and then multiplying the result by the average of the final relative interference levels of the suspected interference periods whose duration is greater than the preset duration threshold; The first factor and the second factor are added to obtain the interference filtering factor.

8. The method for processing liquefied natural gas safety monitoring data according to claim 1, characterized in that: Obtaining the step size of the filter according to the interference filtering factor includes: Normalizing the interference filtering factor to obtain an interference filtering normalized value, and multiplying the interference filtering normalized value by a preset weight to obtain a step size adjustment value; The step size of the filter is obtained by subtracting the step size adjustment value from the preset initial step size.

Images