Supercharging control method, system and equipment for LNG (Liquefied Natural Gas) storage tank

Abstract

The invention relates to the technical field of LNG storage tanks, in particular to an LNG storage tank pressurization control method, system and equipment, and the method comprises the steps that before the injection moment, the optimal time lag is determined by analyzing the correlation degree between pressure data and temperature data at all moments; the temperature influence coefficient of the storage tank is determined by combining the difference of the pressure data between all the adjacent moments and the change trend of the difference between the temperature data; determining the temperature weight and the pressurization weight of the storage tank by analyzing the difference of the pressure data between the storage tank and the pressurizer in the change trend and combining the temperature influence coefficient, and determining the pressure fluctuation index of the storage tank by analyzing the increase rate of the pressure data in the storage tank; and performing pressurization control on the LNG storage tank based on the pressure fluctuation index, the temperature weight and the pressurization weight. The control precision of pressurization of the storage tank is improved, and energy consumption and pollutant emission are reduced.

Description

Technical Field

[0001] The present application relates to the technical field of LNG storage tanks, and in particular to a method, system and equipment for pressurization control of LNG storage tanks. Background Art

[0002] As an efficient and clean form of energy, liquefied natural gas (LNG) is becoming increasingly important in the international energy market. LNG needs to be stored at extremely low temperatures (around -162.5°C), and for ease of transportation and use, it is usually stored in specially designed cryogenic tanks. When the LNG tank supplies gas to the pipeline network, the pressure in the tank will gradually decrease. When the pressure in the tank drops to a certain pressure value, it will affect the speed and capacity of the gas supply. At this time, it is necessary to pressurize the LNG tank through a booster to maintain normal gas supply. The existing technology generally uses a two-position control method to control the pressure of the LNG tank, that is, setting the upper and lower limits of the tank pressure. When the tank pressure is less than the set lower limit, the booster is turned on for boosting. When the tank pressure is greater than the set upper limit, the booster is turned off and the pressure relief valve is opened, thereby leaking some gas, so that the tank pressure is within the set upper and lower limits, thereby achieving control of the tank pressure. However, since the LNG tank cannot be absolutely insulated, the external temperature will also have a certain impact on the LNG tank, causing the tank pressure to rise.

[0003] In the process of pressurization control of LNG storage tanks, the prior art does not take into account the hysteresis effect of temperature on the tank pressure, resulting in the problem of inappropriate pressurization stop time. If the pressurization is stopped too late, it will lead to excessive gasification of LNG, which will not only waste energy but also harm the air quality; if the pressurization is stopped too early, it will lead to insufficient tank pressure. When a large amount of gas supply is required, frequent pressurization is required, resulting in equipment loss. Therefore, the prior art does not take into account the hysteresis effect of temperature on the tank pressure, which reduces the control accuracy of LNG tank pressurization, resulting in increased energy consumption and pollutant emissions. Summary of the invention

[0004] In a first aspect, an embodiment of the present application provides a method for controlling pressurization of an LNG storage tank, the method comprising the following steps: When liquefied natural gas (LNG) starts to flow out of the tank, data collection is performed to collect real-time pressure and liquid level data inside the tank, temperature data outside the tank, and pressure data inside the booster; Extract the mutation points of the pressure data at all times in the tank, and take the time corresponding to the last mutation point in the time series as the injection time; Before the injection time, the correlation between the partial pressure and temperature data is used to determine the optimal time lag, and the temperature influence coefficient of the storage tank is determined by combining the change trend of the pressure data difference and the temperature data difference between all adjacent moments; After the injection time, the pressure data in the tank within the preset period is combined into an internal pressure growth sequence, and characteristic temperature data is extracted from all temperature data in combination with the optimal time lag and the injection time; Compare the differences between the distribution of each characteristic temperature data and all temperature data before the injection moment, correct all elements in the internal pressure growth sequence, and form an internal pressure correction sequence; Based on the changing trend of the differences between all adjacent elements in the internal pressure correction sequence, and the differences between the pressure data of the supercharger and all elements in the internal pressure correction sequence, combined with the temperature influence coefficient, the temperature weight and the boost weight of the storage tank are determined; Based on the changing trend of the pressure data difference between all adjacent moments in the tank after the injection moment, the pressure fluctuation index of the tank is determined; Based on the pressure fluctuation index, temperature weight and boost weight, the boost pressure control of the LNG storage tank is carried out.

[0005] Preferably, the method for determining the optimal time lag is: Before the injection time, the median of the pressure data at all times inside the tank is taken as the upper limit of the range of the time lag k, denoted as a. , the value of k is All integers in ; Before the injection time, the temperature data at all times are sequentially shifted backward in time sequence according to the time lag k, and the correlation coefficient between the result after the shift and the temporal intersection of the pressure data at all times is calculated, which is recorded as the correlation value under the time lag k; Among the correlation values ​​under all time lags, the time lag corresponding to the largest correlation value is taken as the optimal time lag.

[0006] Preferably, the method for determining the temperature influence coefficient of the storage tank is: The first-order difference sequence of the pressure data and the first-order difference sequence of the temperature data at all times inside the storage tank before the injection time are respectively obtained, and a preset number of elements are removed from the first-order difference sequence of the pressure data and the first-order difference sequence of the temperature data, and the remaining elements respectively constitute the internal pressure change sequence and the temperature change sequence; Calculate the ratio between all elements in the internal pressure change sequence and all elements in the temperature change sequence, and take the average of all the ratios as the temperature-pressure ratio index of the storage tank; Temperature influence coefficient of storage tank The expression is: ; In the formula, Indicates the temperature-pressure ratio index of the storage tank; Represents the correlation coefficient between the pressure change series and the temperature change series; represents the optimal time lag.

[0007] Preferably, the process of acquiring the characteristic temperature data is: The time corresponding to the last element in the internal pressure growth sequence is taken as the reference time, and the time at which The temperature data is taken as the characteristic temperature data, where w represents the reference time, z represents the injection time, represents the optimal time lag.

[0008] Preferably, the process of correcting all elements in the internal pressure growth sequence is: If there is characteristic temperature data at the time corresponding to element i in the internal pressure growth sequence, the correction value of element i is The expression is: ; In the formula, Represents the value of element i in the internal pressure growth sequence; Represents the characteristic temperature data of element i at the corresponding moment in the internal pressure growth sequence; Represents the mean of the temperature data at all times before the injection time; represents the temperature proportional index of the storage tank; norm() represents the normalization function; If there is no characteristic temperature data at the time corresponding to element i in the internal pressure growth sequence, the correction value of element i is The expression is: .

[0009] Preferably, the method for determining the temperature weight and the pressure weight of the storage tank is: Obtain the first-order difference sequence of all elements in the internal pressure correction sequence, and use the variance of all elements in the first-order difference sequence as the contribution difference of the storage tank; Calculate the difference between the pressure data at all times inside the supercharger and all elements in the internal pressure correction sequence as the pressure difference of the tank; The ratio of the contribution difference of the storage tank to the pressure difference is used as the pressure increase influence coefficient of the storage tank; The sum of the pressurization influence coefficient and the temperature influence coefficient of the storage tank is calculated, and the ratio of the temperature influence coefficient to the sum, and the ratio of the pressurization influence coefficient to the sum, are recorded as the temperature weight and the pressurization weight, respectively.

[0010] Preferably, the method for determining the pressure fluctuation index of the storage tank is: The first-order difference sequence of the pressure data inside the tank at all times after the injection time is obtained, and the variance of all elements in the first-order difference sequence is used as the pressure fluctuation index of the tank.

[0011] Preferably, the pressurization control of the LNG storage tank includes: When liquefied natural gas LNG flows out of the storage tank, the pressure data, liquid level data, temperature data inside the storage tank and the pressure data inside the booster at all times before the t-th time are respectively formed into an internal pressure cutoff sequence, a liquid level cutoff sequence, a temperature cutoff sequence and a pressure cutoff sequence; The internal pressure cut-off sequence, liquid level cut-off sequence, temperature cut-off sequence and pressure cut-off sequence, as well as the temperature weight, pressurization weight and pressure fluctuation index of the storage tank are used as the input of the neural network, and the pressure data is output as the predicted pressure data at the corresponding time when the injection time is shifted backward to the time t; The deviation between the predicted pressure data at the jth moment after the injection moment and the preset upper limit of the tank pressure is recorded as the upper limit difference. If the upper limit difference is less than the preset threshold, the input of liquefied natural gas LNG into the booster is stopped at the jth moment. Otherwise, the input of liquefied natural gas LNG into the booster continues.

[0012] In a second aspect, an embodiment of the present application provides an LNG storage tank pressurization control system, the system comprising: The tank data acquisition module is used to collect data when liquefied natural gas (LNG) starts to flow out of the tank, and collects real-time pressure and liquid level data inside the tank, temperature data outside the tank, and pressure data inside the booster; The tank weight acquisition module is used to extract the mutation points in the pressure data of the tank at all times, and take the time corresponding to the last mutation point in the time series as the injection time; Before the injection time, the optimal time lag is determined by analyzing the correlation between pressure and temperature data, and the temperature influence coefficient of the storage tank is determined by combining the change trend of the pressure data difference and temperature data difference between all adjacent moments; After the injection time, the pressure data in the tank within the preset period is combined into an internal pressure growth sequence, and characteristic temperature data is extracted from all temperature data in combination with the optimal time lag and the injection time; Compare the differences between the distribution of each characteristic temperature data and all temperature data before the injection moment, correct all elements in the internal pressure growth sequence, and form an internal pressure correction sequence; Based on the variation trend of the differences between all adjacent elements in the internal pressure correction sequence, and the differences between the pressure data of the supercharger and all elements in the internal pressure correction sequence, and in combination with the temperature influence coefficient, the temperature weight and the boost weight of the storage tank are determined; Based on the changing trend of the pressure data difference between all adjacent moments in the tank after the injection moment, the pressure fluctuation index of the tank is determined; The tank boost control module is used to control the boost pressure of the LNG tank based on the pressure fluctuation index, temperature weight and boost weight.

[0013] In the third aspect, an embodiment of the present application also provides an LNG tank pressurization control device, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the steps of any one of the above-mentioned LNG tank pressurization control methods are implemented.

[0014] It can be seen from the above embodiments that the LNG storage tank pressurization control method provided in the embodiments of the present application has at least the following beneficial effects: This application constructs a temperature influence coefficient by analyzing the consistency of temperature and tank pressure changes, and quantifying the degree of change in the tank pressure caused by temperature changes, which can deeply evaluate the degree of influence of temperature on the tank pressure; further, by analyzing the effect of liquefied natural gas (LNG) gasification on the tank pressure, a supercharger influence coefficient is constructed, which can deeply evaluate the effect of the supercharger gasification effect on the tank pressure under the premise of eliminating the temperature lag effect; further, this application constructs a pressure fluctuation index based on the data characteristics of the pressure rise of the tank during supercharging, so as to evaluate whether the pressure rise rate of the tank is consistent. This application improves the control accuracy of tank supercharging by deeply analyzing the effect of temperature and the gasification effect of the supercharger on the tank pressure, and combines the fluctuation characteristics of the tank pressure rise rate, thereby reducing energy consumption and pollutant emissions. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0016] Figure 1 A flowchart of a method for controlling pressurization of an LNG storage tank provided in accordance with an embodiment of the present application; Figure 2 A schematic diagram of a temperature weight and boost weight extraction process provided in one embodiment of the present application; Figure 3 A block diagram of an LNG tank pressurization control system provided for one embodiment of the present application. DETAILED DESCRIPTION

[0017] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of the LNG storage tank pressurization control method, system and equipment proposed in the present application, its specific implementation method, structure, features and effects in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0018] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0019] The specific scheme of a LNG storage tank pressurization control method, system and equipment provided by the present application is described in detail below with reference to the accompanying drawings.

[0020] See also Figure 1 , which shows a flowchart of a method for controlling LNG tank pressurization provided by an embodiment of the present application, the method comprising the following steps: S1: Data collection is performed when liquefied natural gas (LNG) starts to flow out of the storage tank, and real-time pressure data inside the storage tank, liquid level data, temperature data outside the storage tank, and pressure data inside the booster are collected.

[0021] Since LNG storage tanks are usually equipped with pressure sensors and liquid level sensors, this embodiment collects the pressure and liquid level data inside the LNG storage tank through pressure and liquid level sensors respectively, collects ambient temperature data through the temperature sensor outside the LNG storage tank, and collects pressure data inside the supercharger through the pressure sensor.

[0022] Data collection is performed when liquefied natural gas (LNG) begins to flow out of the storage tank. Real-time data is collected on the pressure data inside the storage tank, the liquid level data, the temperature data outside the storage tank, and the pressure data inside the booster. The collection frequency is f. Data collection is stopped when the pressure inside the storage tank reaches the upper limit.

[0023] It should be noted that the value of the acquisition frequency f is artificially set. In this embodiment, the value of the acquisition frequency f is 1 Hz. The implementer can also set it by himself according to the specific situation. This embodiment does not impose any special restrictions.

[0024] In order to eliminate the impact of dimensions between data, all collected data are normalized to obtain processed real-time pressure data, liquid level data and ambient temperature data.

[0025] It should be noted that there are many commonly used normalization methods. In this embodiment, the maximum and minimum normalization method is used to normalize the data. In actual application, as other implementation methods, the implementer can also use the z-score normalization method to normalize the data. Regarding the selection of the normalization method, this embodiment does not make any special restrictions.

[0026] Among them, the maximum and minimum value normalization method is a well-known technology, and the process of normalizing the data is not repeated here.

[0027] S2: Extract the mutation points in the pressure data at all times inside the tank, and take the moment corresponding to the last mutation point in the time series as the injection moment; before the injection moment, determine the optimal time lag by analyzing the correlation between the pressure data and the temperature data at all times, and determine the temperature influence coefficient of the tank by combining the difference in pressure data between all adjacent moments and the changing trend of the difference in temperature data.

[0028] Although LNG storage tanks have a strong thermal insulation effect, the temperature of the LNG inside the tank is extremely low and the tank cannot be completely insulated, so it is still affected by the external ambient temperature. The higher the external ambient temperature, the faster the pressure inside the tank rises.

[0029] When liquefied natural gas (LNG) is gasified in the booster and re-injected into the tank, the pressure inside the tank will rise rapidly and continuously. Before the gaseous natural gas is injected, the impact of ambient temperature on the tank pressure is most significant. Therefore, the impact of ambient temperature on the internal pressure of the tank can be analyzed by analyzing the changes in the tank pressure before the gaseous natural gas is injected.

[0030] Since the effect of ambient temperature on the tank pressure is far less significant than that of gaseous natural gas injection, the internal pressure of the tank has significant stage changes before and after the gaseous natural gas injection. And since the pressure inside the tank will continue to rise after the gaseous natural gas is injected into the tank, there will be no significant mutation.

[0031] Therefore, when liquefied natural gas LNG starts to flow out of the storage tank, a mutation point detection algorithm is used to obtain mutation points in the pressure data inside the storage tank at all times, and the moment corresponding to the last mutation point in the time series is taken as the injection time.

[0032] By analyzing the fluctuation of the pressure data in the tank before the injection time, the influence of the ambient temperature on the tank pressure can be obtained; at the same time, the injection time also reflects the pressure increase process inside the tank from the time when the liquefied natural gas LNG flows out of the tank until the injection time.

[0033] It should be noted that there are many commonly used mutation point detection algorithms. In this embodiment, the PELT algorithm is used to obtain the mutation points in the pressure data. In actual application, as other implementation methods, the implementer can also use the non-stationary time series mutation detection BG algorithm. Regarding the selection of the mutation point detection algorithm, this embodiment does not impose any special restrictions.

[0034] Among them, the PELT algorithm is a well-known technology, and the specific process of detecting mutation points in data will not be described in detail.

[0035] Furthermore, considering that the ambient temperature will affect the tank pressure, the temperature effect has a lag. Therefore, when the ambient temperature changes, the tank pressure will not change immediately, but there is a certain delay. Therefore, it is necessary to first analyze the delay degree of the ambient temperature change on the tank pressure, and then accurately quantify the impact of the ambient temperature on the tank pressure. The specific analysis process of the delay degree of the ambient temperature change on the tank pressure is as follows: Before the injection time, the median of the pressure data at all times inside the tank is taken as the upper limit of the range of the time lag k, denoted as a. , the value of k is All integers in ; Before the injection time, the temperature data at all times are sequentially shifted backward in time sequence according to the time lag k, and the correlation coefficient between the result after the shift and the temporal intersection of the pressure data at all times is calculated, which is recorded as the correlation value under the time lag k; Among all the correlation values ​​under time lag, the time lag corresponding to the largest correlation value is taken as the optimal time lag. The optimal time lag reflects the time delay of the impact of temperature change on the tank pressure. The smaller the optimal time lag, the more direct the impact of temperature on the tank pressure and the greater the impact.

[0036] It should be noted that there are many methods for calculating the correlation coefficient. In this embodiment, the cross-correlation coefficient between the result after translation and the temporal intersection of the pressure data at all moments is used as the correlation coefficient between the result after translation and the temporal intersection of the pressure data at all moments. In actual application, as other implementation methods, the implementer may also adopt other correlation coefficient calculation methods. Regarding the selection of the correlation coefficient calculation method, this embodiment does not impose any special restrictions.

[0037] The calculation method of the cross-correlation coefficient is a well-known technique, and the specific calculation process will not be described in detail.

[0038] Furthermore, in order to eliminate the time delay effect of ambient temperature change on the tank pressure, the first-order difference sequence of the pressure data and the first-order difference sequence of the temperature data at all times inside the tank before the injection time are obtained respectively, and the first preset number of elements are removed from the first-order difference sequence of the pressure data and the first-order difference sequence of the temperature data respectively, and the remaining elements constitute the internal pressure change sequence and the temperature change sequence respectively; It should be noted that the value of the preset number in this embodiment is the same as the value of the optimal time lag.

[0039] The process of obtaining the first-order difference sequence is a well-known technique, and the specific steps of obtaining the first-order difference sequence are not described in detail.

[0040] Further, the ratios between all elements in the internal pressure change sequence and all elements in the temperature change sequence are calculated, and the average of all the ratios is used as the temperature-pressure ratio index of the storage tank; Temperature influence coefficient of storage tank The expression is: ; In the formula, Indicates the temperature-pressure ratio index of the storage tank; Represents the correlation coefficient between the pressure change series and the temperature change series; represents the optimal time lag.

[0041] It should be noted that there are many methods for measuring the degree of correlation between sequences. In this embodiment, the absolute value of the Pearson correlation coefficient between the pressure change sequence and the temperature change sequence is used as the correlation coefficient between the pressure change sequence and the temperature change sequence. In actual application, as other implementation methods, the implementer may also use the Spearman correlation coefficient or the Kendall rank correlation coefficient. Regarding the selection of the method for measuring the degree of correlation between sequences, this embodiment does not impose any special restrictions.

[0042] The calculation method of the Pearson correlation coefficient is a well-known technique, and the specific calculation process will not be described in detail.

[0043] According to the temperature influence coefficient of the storage tank, it can be understood that if the temperature-pressure proportional index is larger, the correlation coefficient between the pressure change sequence and the temperature change sequence is smaller, then the obtained temperature influence coefficient is larger, indicating that with the change of temperature, the degree of change of the storage tank pressure is greater, and the response of the storage tank pressure is faster; conversely, if the temperature-pressure proportional index is smaller, the correlation coefficient between the pressure change sequence and the temperature change sequence is larger, then the obtained temperature influence coefficient is smaller, indicating that with the change of temperature, the degree of change of the storage tank pressure is smaller, and the response of the storage tank pressure is slower.

[0044] S3: Analyze the difference in the changing trend of the pressure data between the storage tank and the booster, as well as the growth rate of the pressure data inside the storage tank, and determine the temperature weight, boost weight and pressure fluctuation index of the storage tank respectively.

[0045] Since the pressurization process of the storage tank is to let liquefied natural gas (LNG) flow into the supercharger, then gasify it into gaseous natural gas and inject it into the storage tank to achieve pressurization, it has a direct impact on the tank pressure. Therefore, it is also necessary to analyze the impact of the supercharger's gasification effect on the tank pressure, so as to achieve subsequent precise pressurization control of the tank. The specific process is: S301: After the injection time, all pressure data inside the storage tank within a preset period are combined into an internal pressure growth sequence, and characteristic temperature data are extracted from the temperature data at all times in combination with the optimal time lag and the injection time.

[0046] After the injection moment, all the pressure data inside the storage tank within a preset period of time are combined into an internal pressure growth sequence.

[0047] It should be noted that the time length between the moment when the liquefied natural gas LNG starts to flow out of the storage tank and the injection moment is counted, and the time length is used as the value of the preset time period length.

[0048] It should be understood that when the tank pressure drops to the lower pressure limit, pressurization begins. Since the tank is very large, pressurization takes a long time. Therefore, the amount of data collected after the injection time is large. The time from the start of liquefied natural gas output to the injection time is shorter than the overall pressurization time. Therefore, the data collection time after the injection time is greater than the time between the moment when liquefied natural gas LNG starts to flow out of the tank and the injection time.

[0049] The time corresponding to the last element in the internal pressure growth sequence is taken as the reference time, and the time at which The temperature data is taken as the characteristic temperature data, where w represents the reference time, z represents the injection time, represents the optimal time lag.

[0050] S302: Compare the differences between each characteristic temperature data and the average distribution of all temperature data before the injection moment, correct all elements in the internal pressure growth sequence, and the correction results constitute the internal pressure correction sequence.

[0051] If there is characteristic temperature data at the time corresponding to element i in the internal pressure growth sequence, the correction value of element i is The expression is: ; In the formula, Represents the value of element i in the internal pressure growth sequence; Represents the characteristic temperature data of element i at the corresponding moment in the internal pressure growth sequence; Represents the mean of the temperature data at all times before the injection time; represents the temperature proportional index of the storage tank; norm() represents the normalization function; If there is no characteristic temperature data at the time corresponding to element i in the internal pressure growth sequence, the correction value of element i is The expression is: .

[0052] According to the correction value of each element in the internal pressure growth sequence, it can be understood that if the difference between the characteristic temperature data and the mean value of the temperature data before the injection moment is larger, the temperature proportional index is larger, which means that the temperature difference has a greater impact on the change of the tank pressure. Therefore, the tank pressure is corrected by the temperature difference; conversely, if the difference between the characteristic temperature data and the mean value of the temperature data before the injection moment is smaller, the temperature proportional index is smaller, which means that the temperature difference has a smaller impact on the change of the tank pressure.

[0053] By calculating the difference in the mean temperature between the characteristic temperature data and the mean of the temperature data at all times before the injection moment, the change in the tank pressure corresponding to the temperature difference is obtained, and the degree of change caused by the temperature change is subtracted from the existing tank pressure, thereby removing the change in the tank pressure caused by the temperature change.

[0054] S303: Based on the changing trend of the differences between all adjacent elements in the internal pressure correction sequence, and the differences between the pressure data of the supercharger at all times and all elements in the internal pressure correction sequence, and in combination with the temperature influence coefficient, the temperature weight and the boost weight of the storage tank are determined.

[0055] Since the speed at which LNG flows out of the storage tank is uniform, the volume of LNG entering the supercharger at each moment is also consistent. Therefore, if the gasification rate of the supercharger for LNG is consistent, the increase in the volume of LNG flowing out at each moment should also be consistent with the increase in the tank pressure.

[0056] The first-order difference sequence of all elements in the internal pressure correction sequence is obtained, and the variance of all elements in the first-order difference sequence is used as the contribution difference of the storage tank; since liquefied natural gas LNG flows out at a uniform speed, the greater the contribution difference, the greater the possibility that the inconsistent pressure growth of the storage tank will be caused by the inconsistent gasification efficiency of the booster.

[0057] Furthermore, since the pressure inside the supercharger will rise after the liquefied natural gas LNG is gasified, the internal pressure of the supercharger can also reflect the degree of gasification of the liquefied natural gas LNG by the supercharger. At the same time, since there is a time delay between the gasification of the liquefied natural gas LNG by the supercharger and the injection of the gaseous natural gas into the storage tank, the influence of the supercharger gasification change on the storage tank pressure is analyzed by analyzing the difference between the pressure data of the supercharger at all times and all elements in the internal pressure correction sequence, specifically: Calculate the difference between the pressure data at all times inside the supercharger and all elements in the internal pressure correction sequence as the pressure difference of the tank; It should be noted that there are many methods for measuring the differences between data groups. In this embodiment, the dynamic time warping (DTW) distance between the corrected values ​​of all elements in the internal pressure growth sequence and all pressure data is used as the difference between the corrected values ​​of all elements in the internal pressure growth sequence and all pressure data; in actual application, as other implementation methods, the implementer may also adopt other methods for measuring the differences between data groups, such as Euclidean distance or Manhattan distance. Regarding the selection of methods for measuring the differences between data groups, this embodiment does not impose any special restrictions.

[0058] The calculation method of the dynamic time warping (DTW) distance is a well-known technology, and its specific calculation process will not be repeated here.

[0059] Furthermore, the ratio of the contribution difference of the storage tank to the pressure difference is used as the pressurization influence coefficient of the storage tank.

[0060] According to the supercharger influence coefficient of the storage tank, it can be understood that the supercharger influence coefficient can reflect the influence of the gasification rate of the supercharger on the storage tank pressure. If the contribution difference is larger and the pressure difference is smaller, the obtained supercharger influence coefficient is larger, which is reflected in the process of tank pressurization, and the influence of supercharger gasification is greater; conversely, if the contribution difference is smaller and the pressure difference is larger, the obtained supercharger influence coefficient is smaller, which is reflected in the process of tank pressurization, and the influence of supercharger gasification is smaller.

[0061] Further, the sum of the pressurization influence coefficient and the temperature influence coefficient of the storage tank is calculated, and the ratio of the temperature influence coefficient to the sum, and the ratio of the pressurization influence coefficient to the sum, are recorded as the temperature weight and the pressurization weight, respectively.

[0062] Preferably, the schematic diagram of the temperature weight and boost weight extraction process provided in this embodiment is as follows: Figure 2 shown.

[0063] S304: Determine a pressure fluctuation index of the storage tank based on a change trend of the pressure data difference between all adjacent moments in the storage tank after the injection moment.

[0064] Furthermore, in order to achieve accurate prediction of the tank pressure at subsequent moments, it is necessary to evaluate the rising rate of the tank pressure after the injection of gaseous natural gas to detect whether its rising rate is relatively consistent and stable.

[0065] The first-order difference sequence of the pressure data inside the tank at all times after the injection is obtained, and the variance of all elements in the first-order difference sequence is used as the pressure fluctuation index of the tank. The pressure fluctuation index can reflect whether the rising rate of the tank pressure is consistent after the gaseous natural gas is injected into the tank.

[0066] Step S4: Based on the pressure fluctuation index, the temperature weight and the boost weight, the boost control is performed on the LNG storage tank.

[0067] The liquefied natural gas LNG starts flowing out of the storage tank and is gasified into gaseous natural gas, and is not injected into the storage tank until the injection time is reached. Therefore, if the liquefied natural gas LNG outflow time is taken as the starting time, and the time between the injection time and the starting time is recorded as h, then the difference between the liquefied natural gas inflow time and the starting time is h. Therefore, by predicting the tank pressure at each moment after h time, and by calculating the difference between the upper limit of the tank pressure setting and the predicted tank pressure, it is possible to evaluate when it is necessary to stop the outflow of LNG from the tank.

[0068] When liquefied natural gas LNG flows out of the storage tank, the pressure data, liquid level data, temperature data inside the storage tank and the pressure data inside the booster at all times before the t-th time are respectively formed into an internal pressure cutoff sequence, a liquid level cutoff sequence, a temperature cutoff sequence and a pressure cutoff sequence; In this embodiment, the internal pressure cut-off sequence, liquid level cut-off sequence, temperature cut-off sequence and pressure cut-off sequence, as well as the temperature weight, pressurization weight and pressure fluctuation index of the storage tank are used as the input of the BP neural network. The input layer has a total of 7 input nodes, and the output layer has only one output node, which outputs pressure data as the predicted pressure data at the corresponding time when the injection time is shifted backward to the time t; Among them, the optimization algorithm of the BP neural network is the SGD algorithm, the loss function is MSE; the division of the training set and the test set is 7:3, the hidden layer is 5 layers, and the number of iterations is 50 times.

[0069] The deviation between the predicted pressure data at the jth moment after the injection moment and the preset upper limit of the tank pressure is recorded as the upper limit difference. If the upper limit difference is less than the preset threshold value, it means that the liquefied natural gas LNG in the booster at this time can meet the pressurization of the subsequent tank pressure. In this case, the input of liquefied natural gas LNG into the booster is stopped at the jth moment. Otherwise, the input of liquefied natural gas LNG into the booster is continued.

[0070] It should be noted that in this embodiment, the lower limit of the LNG storage tank pressure is 0.2MPa and the upper limit is 0.5MPa. Therefore, the value of the preset upper limit of the storage tank pressure in this embodiment is 0.5MPa, and the implementer can also set it by himself according to the specific situation. There is no special restriction in this embodiment. In addition, the value of the preset threshold in this embodiment is also artificially set. The value of the preset threshold in this embodiment is 0.005MPa. The implementer can also set it by himself according to the specific situation. There is no special restriction in this embodiment.

[0071] Based on the same inventive concept as the above method, the embodiment of the present application also provides an LNG storage tank pressurization control system, including: The tank data acquisition module is used to collect data when liquefied natural gas (LNG) starts to flow out of the tank, and collects real-time pressure and liquid level data inside the tank, temperature data outside the tank, and pressure data inside the booster; The tank weight acquisition module is used to extract the mutation points in the pressure data of the tank at all times, and take the time corresponding to the last mutation point in the time series as the injection time; Before the injection time, analyze the correlation between pressure and temperature data, determine the optimal time lag, and determine the temperature influence coefficient of the tank by combining the change trend of the pressure data difference and temperature data difference between all adjacent moments; After the injection time, the pressure data in the tank within the preset period is combined into an internal pressure growth sequence, and characteristic temperature data is extracted from all temperature data in combination with the optimal time lag and the injection time; Compare the differences between each characteristic temperature data and the average distribution of all temperature data before the injection moment, correct all elements in the internal pressure growth sequence, and form an internal pressure correction sequence; Based on the changing trend of the differences between all adjacent elements in the internal pressure correction sequence, and the differences between the pressure data of the supercharger and all elements in the internal pressure correction sequence, combined with the temperature influence coefficient, the temperature weight and the boost weight of the storage tank are determined; Based on the changing trend of the pressure data difference between all adjacent moments in the tank after the injection moment, the pressure fluctuation index of the tank is determined; The tank boost control module is used to control the boost pressure of the LNG tank based on the pressure fluctuation index, temperature weight and boost weight.

[0072] The present application provides a block diagram of a LNG tank pressurization control system, such as Figure 3 shown.

[0073] Based on the same inventive concept as the above method, an embodiment of the present application also provides an LNG tank pressurization control device, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the steps of any one of the above-mentioned LNG tank pressurization control methods are implemented.

[0074] It should be noted that the above sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying 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.

[0075] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0076] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for controlling the pressurization of an LNG storage tank, characterized in that: The method comprises the following steps: When liquefied natural gas (LNG) starts to flow out of the tank, data collection is performed to collect real-time pressure and liquid level data inside the tank, temperature data outside the tank, and pressure data inside the booster; Extract the mutation points of the pressure data at all times in the tank, and take the time corresponding to the last mutation point in the time series as the injection time; Before the injection time, analyze the correlation between pressure and temperature data, determine the optimal time lag, and determine the temperature influence coefficient of the tank by combining the change trend of the pressure data difference and temperature data difference between all adjacent moments; After the injection time, the pressure data in the tank within the preset period is combined into an internal pressure growth sequence, and characteristic temperature data is extracted from all temperature data in combination with the optimal time lag and the injection time; Compare the differences between the distribution of each characteristic temperature data and all temperature data before the injection moment, correct all elements in the internal pressure growth sequence, and form an internal pressure correction sequence; Based on the changing trend of the differences between all adjacent elements in the internal pressure correction sequence, and the differences between the pressure data of the supercharger and all elements in the internal pressure correction sequence, combined with the temperature influence coefficient, the temperature weight and the boost weight of the storage tank are determined; Based on the changing trend of the pressure data difference between all adjacent moments in the tank after the injection moment, the pressure fluctuation index of the tank is determined; Based on the pressure fluctuation index, temperature weight and boost weight, the boost pressure control of the LNG storage tank is carried out.

2. A LNG storage tank pressurization control method according to claim 1, characterized in that: The method for determining the optimal time lag is: Before the injection time, the median of the pressure data at all times inside the tank is taken as the upper limit of the range of the time lag k, denoted as a. , the value of k is All integers in ; Before the injection time, the temperature data at all times are sequentially shifted backward in time sequence according to the time lag k, and the correlation coefficient between the result after the shift and the temporal intersection of the pressure data at all times is calculated, which is recorded as the correlation value under the time lag k; Among the correlation values ​​under all time lags, the time lag corresponding to the largest correlation value is taken as the optimal time lag.

3. The LNG storage tank pressurization control method according to claim 1, characterized in that: The method for determining the temperature influence coefficient of the storage tank is: The first-order difference sequence of the pressure data and the first-order difference sequence of the temperature data at all times inside the storage tank before the injection time are respectively obtained, and a preset number of elements are removed from the first-order difference sequence of the pressure data and the first-order difference sequence of the temperature data, and the remaining elements respectively constitute the internal pressure change sequence and the temperature change sequence; Calculate the ratio between all elements in the internal pressure change sequence and all elements in the temperature change sequence, and use the average of all the ratios as the temperature-pressure ratio index of the storage tank; Temperature influence coefficient of storage tank The expression is: ; In the formula, Indicates the temperature-pressure ratio index of the storage tank; Represents the correlation coefficient between the pressure change series and the temperature change series; represents the optimal time lag.

4. The LNG storage tank pressurization control method according to claim 1, characterized in that: The process of acquiring the characteristic temperature data is as follows: The time corresponding to the last element in the internal pressure growth sequence is taken as the reference time, and the time at which The temperature data is taken as the characteristic temperature data, where w represents the reference time, z represents the injection time, represents the optimal time lag.

5. The LNG storage tank pressurization control method according to claim 3, characterized in that: The process of correcting all elements in the internal pressure growth sequence is: If there is characteristic temperature data at the time corresponding to element i in the internal pressure growth sequence, the correction value of element i is The expression is: ; In the formula, Represents the value of element i in the internal pressure growth sequence; Represents the characteristic temperature data of element i at the corresponding moment in the internal pressure growth sequence; Represents the mean of the temperature data at all times before the injection time; represents the temperature proportional index of the storage tank; norm() represents the normalization function; If there is no characteristic temperature data at the time corresponding to element i in the internal pressure growth sequence, the correction value of element i is The expression is: .

6. The LNG storage tank pressurization control method according to claim 1, characterized in that: The method for determining the temperature weight and the pressure weight of the storage tank is: Obtain the first-order difference sequence of all elements in the internal pressure correction sequence, and use the variance of all elements in the first-order difference sequence as the contribution difference of the storage tank; Calculate the difference between the pressure data at all times inside the supercharger and all elements in the internal pressure correction sequence as the pressure difference of the tank; The ratio of the contribution difference of the storage tank to the pressure difference is used as the pressurization influence coefficient of the storage tank; The sum of the pressurization influence coefficient and the temperature influence coefficient of the storage tank is calculated, and the ratio of the temperature influence coefficient to the sum, and the ratio of the pressurization influence coefficient to the sum, are recorded as the temperature weight and the pressurization weight, respectively.

7. The LNG storage tank pressurization control method according to claim 1, characterized in that: The method for determining the pressure fluctuation index of the storage tank is: The first-order difference sequence of the pressure data inside the tank at all times after the injection time is obtained, and the variance of all elements in the first-order difference sequence is used as the pressure fluctuation index of the tank.

8. The LNG storage tank pressurization control method according to claim 1, characterized in that: The pressurization control of the LNG storage tank includes: When liquefied natural gas LNG flows out of the storage tank, the pressure data, liquid level data, temperature data inside the storage tank and the pressure data inside the booster at all times before the t-th time are respectively formed into an internal pressure cutoff sequence, a liquid level cutoff sequence, a temperature cutoff sequence and a pressure cutoff sequence; The internal pressure cut-off sequence, liquid level cut-off sequence, temperature cut-off sequence and pressure cut-off sequence, as well as the temperature weight, pressurization weight and pressure fluctuation index of the storage tank are used as the input of the neural network, and the pressure data is output as the predicted pressure data at the corresponding time when the injection time is shifted backward to the time t; The deviation between the predicted pressure data at the jth moment after the injection moment and the preset upper limit of the tank pressure is recorded as the upper limit difference. If the upper limit difference is less than the preset threshold, the input of liquefied natural gas LNG into the booster is stopped at the jth moment. Otherwise, the input of liquefied natural gas LNG into the booster continues.

9. An LNG tank pressurization control system, implementing an LNG tank pressurization control method as claimed in claim 1, characterized in that: The system comprises: The tank data acquisition module is used to collect data when liquefied natural gas (LNG) starts to flow out of the tank, and collects real-time pressure and liquid level data inside the tank, temperature data outside the tank, and pressure data inside the booster; The tank weight acquisition module is used to extract the mutation points in the pressure data of the tank at all times, and take the time corresponding to the last mutation point in the time series as the injection time; Before the injection time, the optimal time lag is determined by analyzing the correlation between pressure and temperature data, and the temperature influence coefficient of the storage tank is determined by combining the change trend of the pressure data difference and temperature data difference between all adjacent moments; After the injection time, the pressure data in the tank within the preset period is combined into an internal pressure growth sequence, and characteristic temperature data is extracted from all temperature data in combination with the optimal time lag and the injection time; Compare the differences between each characteristic temperature data and the average distribution of all temperature data before the injection moment, correct all elements in the internal pressure growth sequence, and form an internal pressure correction sequence; Based on the variation trend of the differences between all adjacent elements in the internal pressure correction sequence, and the differences between the pressure data of the supercharger and all elements in the internal pressure correction sequence, and in combination with the temperature influence coefficient, the temperature weight and the boost weight of the storage tank are determined; Based on the changing trend of the pressure data difference between all adjacent moments in the tank after the injection moment, the pressure fluctuation index of the tank is determined; The tank boost control module is used to control the boost pressure of the LNG tank based on the pressure fluctuation index, temperature weight and boost weight.

10. An LNG tank pressurization control device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the LNG storage tank pressurization control method as described in any one of claims 1-8 are implemented.

Images