Natural gas long-distance pipeline supply-demand relation regulation and control method and device

Abstract

The invention relates to the technical field of natural gas regulation and control, in particular to a natural gas long-distance pipeline supply-demand relation regulation and control method and device, and the method comprises the steps: building a multivariate parameter mathematical model between historical influence factor data and gas consumption data according to the obtained historical influence factor data and gas consumption data, namely a pipeline gas consumption prediction model; using the prediction model to predict the gas consumption of each user in the future predetermined time period; and if a deviation exists between the gas consumption prediction result in the predetermined time period and the gas consumption in the actual predetermined time period, determining a deviation index of the prediction result of the pipeline gas consumption prediction model, correcting the prediction result of the pipeline gas consumption prediction model according to the deviation index, and regulating and controlling the real-time gas supply according to the corrected prediction result. The problems that the working efficiency and the regulation and control precision are low, it cannot be guaranteed that all types of users are in the optimal supply and demand state, and safe and stable operation of the pipeline is not facilitated due to the fact that an existing natural gas long-distance pipeline is adjusted through manual experience during gas consumption regulation and control are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of natural gas regulation, and particularly relates to a natural gas long-distance pipeline supply-demand relationship regulation method and device. BACKGROUND

[0002] The downstream gas volume control of the natural gas long-distance pipeline is currently mainly achieved by manual regulation, and thus the gas volume control of downstream users needs to be performed by an operator. The control relies on experience and has no unified control method. The control result is affected by the individual ability level and has subjective differences. The control result cannot be comprehensively monitored and analyzed, and has low guiding significance for actual production. The work efficiency is low, the labor intensity is high, and the control precision is low. The control method is not conducive to the safe, stable and efficient operation of the long-distance pipeline, and cannot meet the needs of the intelligent development of the long-distance pipeline. SUMMARY

[0003] The present application provides a natural gas long-distance pipeline supply-demand relationship regulation method and device to solve the problem that the existing natural gas long-distance pipeline uses manual experience to adjust the gas consumption during the gas consumption regulation, which results in low work efficiency and regulation precision, and cannot guarantee that each type of user is in the best supply-demand state, and is not conducive to the safe and stable operation of the pipeline.

[0004] According to an aspect of the present application, a natural gas long-distance pipeline supply-demand relationship regulation method is provided, which includes:

[0005] acquiring historical influence factor data related to gas consumption prediction and corresponding gas consumption data;

[0006] establishing a multivariate parameter mathematical model between the historical influence factor data and the gas consumption data, the multivariate parameter mathematical model being a pipeline gas volume prediction model;

[0007] predicting the future scheduled period gas consumption of each user by using the pipeline gas volume prediction model, and regulating the real-time gas supply according to the scheduled period gas consumption prediction result;

[0008] determining whether there is a deviation between the scheduled period gas consumption prediction result and the actual scheduled period gas consumption. If yes, a deviation index of the pipeline gas volume prediction model prediction result is determined, the prediction result of the pipeline gas volume prediction model is corrected according to the deviation index, and the real-time gas supply is regulated according to the corrected prediction result.

[0009] Preferably, before the historical influence factor data related to gas consumption prediction is acquired, the historical influence factor data related to gas consumption prediction is determined by a method including:

[0010] Obtaining historical pipeline operation parameter data, performing data dimensionality reduction analysis on the pipeline operation parameter data to obtain historical influence factor data related to gas consumption prediction.

[0011] Preferably, the influence factor data at least includes:

[0012] Each user flow, each user hourly gas consumption, each user daily cumulative gas consumption, each user annual cumulative gas consumption, pipeline natural gas flow rate, pipeline import temperature, pipeline export temperature, pipeline temperature change, pipeline import pressure, pipeline export pressure, pipeline import and export pressure difference, pipeline pressure change, average ambient temperature, maximum temperature and minimum temperature.

[0013] Preferably, the method for predicting future scheduled period gas consumption of each user by using the pipeline gas consumption prediction model comprises:

[0014] Predicting total gas consumption of each user in a future scheduled number of hours by using the pipeline gas consumption prediction model;

[0015] According to historical gas consumption data, determining the proportion of gas consumption of each hour in the scheduled number of hours;

[0016] According to the proportion of gas consumption and the total gas consumption, the future scheduled period gas consumption.

[0017] Preferably, the method for determining the proportion of gas consumption of each hour in the scheduled number of hours according to historical gas consumption data comprises:

[0018] Finding historical same-period gas consumption proportion data corresponding to the hour for which the proportion of gas consumption is to be determined;

[0019] Respectively determining the average value of the proportion of gas consumption corresponding to the same hour as the hour for which the proportion of gas consumption is to be determined in the previous scheduled different number of days of the date of the historical same-period gas consumption proportion data;

[0020] Comparing the obtained average value of each proportion with the corresponding historical same-period gas consumption proportion respectively to determine the correlation coefficient therebetween, and the scheduled different number of days corresponding to the maximum value among all correlation coefficients is the optimal sampling period;

[0021] The proportion of gas consumption of each hour in the scheduled number of hours is the average value of the proportion of gas consumption of the same hour corresponding to each hour in the previous optimal sampling period of the date of each hour.

[0022] Preferably, the method for determining the deviation index of the prediction result of the pipeline gas consumption prediction model comprises:

[0023] Determining the deviation index by using formula (1);

[0024]

[0025] F = (a + β1×P + β2×K)×SP 预 F is the predicted remaining total weather amount, 实 F is the actual remaining total weather amount, F is the predicted total weather amount except for the occurred hour.

[0026] Preferably, the method for correcting the prediction result of the pipeline gas amount prediction model according to the deviation index comprises:

[0027] According to the deviation index, the prediction result of the pipeline gas amount prediction model is corrected by using formula (2);

[0028] SP 目标 = (a + β1×P + β2×K)×SP 预测 (2);

[0029] SP 目标 is the corrected prediction result, a, β1, β2 are correction coefficients, P = P1 / P2, wherein P1 is the inlet pressure of the regulating valve, P2 is the outlet pressure of the regulating valve, SP 预测 is the prediction result of the pipeline gas amount prediction model.

[0030] Preferably, it further comprises: inputting the newly generated user influence factor data and the corresponding actual gas consumption data into the pipeline gas amount prediction model at every interval of a predetermined time to train the model.

[0031] Preferably, the method for regulating the real-time gas supply amount comprises:

[0032] In the case of insufficient natural gas, uninterrupted gas supply is provided to the first-level users among all users according to the gas consumption in the corresponding prediction result, and intermittent gas supply is provided to the second-level users.

[0033] If the gas consumption of a user in the future predetermined period exceeds a predetermined value in the prediction result, the pre-stored natural gas in the gas pipeline is consumed in the future predetermined period of the user.

[0034] According to an aspect of the present application, a natural gas long-distance pipeline supply-demand relationship control device is provided, comprising:

[0035] An acquisition unit is configured to acquire historical influence factor data related to gas consumption prediction and corresponding gas consumption data;

[0036] A model establishing unit is configured to establish a multivariate parameter mathematical model between the historical influence factor data and the gas consumption data, wherein the multivariate parameter mathematical model is a pipeline gas amount prediction model.

[0037] a prediction unit configured to predict the gas consumption of each user in a future scheduled period by using the pipeline gas volume prediction model, and to regulate the real-time gas supply according to the prediction result of the scheduled period gas consumption;

[0038] a correction unit configured to determine whether there is a deviation between the prediction result of the scheduled period gas consumption and the actual scheduled period gas consumption, and if yes, to determine a deviation index of the prediction result of the pipeline gas volume prediction model, to correct the prediction result of the pipeline gas volume prediction model according to the deviation index, and to regulate the real-time gas supply according to the corrected prediction result.

[0039] The present application has at least the following beneficial effects:

[0040] The present application provides a natural gas long-distance pipeline supply-demand relationship regulation method and device, which establishes a pipeline gas volume prediction model by using the influence factor data and the gas consumption data in historical data, thereby predicting the gas consumption of each user in each period in the future, automatically regulating the global pipeline supply-demand relationship according to the prediction result, solving the problem of mismatch between the supply and demand of the natural gas long-distance pipeline network, guaranteeing reliable gas supply for downstream users, and improving the control precision and work efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0041] The accompanying drawings, which are incorporated herein and form part of the specification, illustrate embodiments consistent with the present application and, together with the description, further serve to explain the principles of the application.

[0042] Figure 1 A flow chart of a natural gas long-distance pipeline supply-demand relationship regulation method according to an embodiment of the present application is shown;

[0043] Figure 2 A fitting curve diagram of predicted gas consumption and actual gas consumption according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0044] Various exemplary embodiments, features and aspects of the present application will be described in detail below with reference to the accompanying drawings. The same reference numbers in the drawings represent functionally similar or the same elements. Although various aspects of the embodiments are shown in the drawings, the drawings are not necessarily drawn to scale unless specifically indicated.

[0045] The word "exemplary" is used herein to mean "serving as an example, instance, or illustration." Any implementation described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other implementations.

[0046] The term "and / or", used in the present document, only describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which can represent the three cases of A alone, A and B together, and B alone. In addition, the term "at least one" in the present document means any one of a plurality or any combination of at least two of a plurality, for example, at least one of A, B and C includes any one or more elements selected from the set consisting of A, B and C.

[0047] In addition, in order to better illustrate the present application, numerous specific details are given in the specific embodiments below. Those skilled in the art should understand that the present application can also be implemented without certain specific details. In some examples, methods, means, elements and circuits well known to those skilled in the art are not described in detail, in order to highlight the main idea of the present application.

[0048] Figure 1 A flow chart of a natural gas long-distance pipeline supply-demand relationship regulation method according to an embodiment of the present application is shown; Figure 2 A fitting curve diagram of predicted gas consumption and actual gas consumption according to an embodiment of the present application is shown. As Figures 1-2 The natural gas long-distance pipeline supply-demand relationship regulation method includes: step S01: obtaining historical influence factor data related to gas consumption prediction and corresponding gas consumption data; step S02: establishing a multivariate parameter mathematical model between the historical influence factor data and the gas consumption data, the multivariate parameter mathematical model being a pipeline gas consumption prediction model; step S03: predicting future scheduled period gas consumption of each user by using the pipeline gas consumption prediction model, and regulating real-time gas supply according to the scheduled period gas consumption prediction result; step S04: determining whether there is a deviation between the scheduled period gas consumption prediction result and actual scheduled period gas consumption, and if yes, determining a deviation index of the pipeline gas consumption prediction model prediction result, correcting the pipeline gas consumption prediction model prediction result according to the deviation index, and regulating real-time gas supply according to the corrected prediction result.

[0049] The natural gas long-distance pipeline supply-demand relationship regulation method provided by the embodiment of the present application specifically includes the following steps:

[0050] Step S01: obtaining historical influence factor data related to gas consumption prediction and corresponding gas consumption data.

[0051] In the present application, before the historical influence factor data related to gas consumption prediction is obtained, the historical influence factor data related to gas consumption prediction is determined, and the method includes: obtaining historical pipeline operation parameter data, performing data dimensionality reduction analysis on the pipeline operation parameter data, and obtaining the historical influence factor data related to gas consumption prediction.

[0052] In the embodiment of the present invention, the actual gas consumption of users is affected by many factors in the actual physical world. If one wants to express it with a mathematical function, it must be a multivariate complex function. Each influencing factor, that is, each parameter in the pipeline operation, is regarded as a dimension of the physical world. When it is expressed with one or more variables, the solution, calculation, and correction of the function are bound to be time-consuming and laborious. However, in the actual production process, due to the certain gas storage capacity of the long-distance pipeline, the unit of measurement when supplying gas does not need to be accurate to 1m. 3 , usually the acceptable accuracy range is ±100m 3 / h. In order to predict the gas consumption of downstream users with the accuracy allowed, the main factors affecting gas consumption should be found, and the influence of secondary factors should be simplified or discarded. Therefore, a data dimensionality reduction analysis model is established to perform data dimensionality reduction analysis on various parameters. The model takes into account all possible factors that can affect the gas consumption of users, such as natural factors such as temperature and holidays, as well as human factors. Then, a fitting function is generated to evaluate the influencing factors one by one. When the correction coefficient of the factor is usually less than 1×10 -5 When the gas volume is within ±1m 3 , the influence of this factor can be ignored. This model can simplify the algorithm and improve the operation speed without affecting the accuracy.

[0053] In order to predict the gas consumption of downstream users, the main factors affecting gas consumption should be identified, and the influence of secondary factors should be simplified or discarded to achieve the goal of simplifying the algorithm and improving the operation speed without affecting the accuracy. Among them, the dimensionality reduction method that can be used is the existing dimensionality reduction model. The existing commonly used dimensionality reduction analysis method can be used.

[0054] In the present invention, the influencing factor data include at least: the flow rate of each user, the hourly gas volume of each user, the cumulative gas volume of each user throughout the day, the annual cumulative gas volume of each user, the natural gas flow rate in the pipeline, the pipeline inlet temperature, the pipeline outlet temperature, the pipeline temperature change, the pipeline inlet pressure, the pipeline outlet pressure, the pipeline inlet and outlet pressure difference, the pipeline pressure change, the average ambient temperature, the maximum temperature and the minimum temperature.

[0055] In the embodiment of the present invention, pipeline operation parameter data mainly refers to the parameter values ​​related to the long-distance pipeline network. Among them, the more important influencing factors (parameters) related to the prediction include but are not limited to parameter data such as annual, daily and hourly gas consumption, flow, temperature, pressure, etc. of each user. These parameter data include historical data as well as real-time data.

[0056] The historical data (historical influence factor data and corresponding gas consumption data) used to predict the gas consumption of a certain user at a certain period is the historical same period data of the user. The historical same period data is: if the gas consumption of a certain user on a certain day in a certain month is to be predicted, the corresponding historical same period data used is the historical same period data of the same month and day and / or the previous N days of the user in the previous year or years.

[0057] Step S02: establishing a multiple parameter mathematical model between the historical influence factor data and the gas consumption data, which is a pipeline gas quantity prediction model.

[0058] In the embodiments of the present application, MATLAB mathematical software can be used to establish the prediction model. Linear regression analysis method is used to fit the independent variables and dependent variables when establishing the prediction model. Linear regression is a data processing method. In the case where the precision is allowed, a simple function is used to fit a complex function to simplify the calculation method and improve the operation speed.

[0059] Taking a certain type of industrial user as an example, the daily gas quantity prediction model is described. The historical data required to establish the prediction model is the total sample data, such as including the pipeline operation parameter data in 2019, 2020 and 2021; wherein each influence factor data is an independent variable, such as X1, X2, X3. The process of predicting the daily gas quantity is:

[0060] The independent variable X1 is selected, mainly considering the temperature factor. According to the daily average temperature, the corresponding gas quantity of the day within ±1℃ of the average ambient temperature in the past three years is selected. After aggregating by year, taking the average value, the data is processed in the form of an arithmetic sequence, that is, an arithmetic sequence is established using the data in 2019 and 2020 to calculate the data in 2021, then the average value of the data in 2021 and the calculated data is determined as the value of the independent variable X1, and the data is calculated recursively to obtain the X1 sequence.

[0061] The mathematical expression is:

[0062] First, the average value of the gas quantity Q in all days within ±1℃ of a certain temperature m℃, i.e. a certain temperature, in the previous 2019 and 2020 is calculated;

[0063]

[0064] The average value of the gas quantity at the unified temperature in 2021 is calculated according to the average value of the samples of the same temperature (ambient temperature) in 2019 and 2020;

[0065]

[0066] The average value is calculated according to the average flow rate derived in 2021 and the actual gas quantity in 2021;

[0067]

[0068] Select the independent variable X2, mainly consider the recent gas transmission amount factor, the average value of the previous three days of gas transmission amount, the data is calculated by recursion, and the X2 sequence is obtained.

[0069] The mathematical expression is:

[0070]

[0071] Select the independent variable X3, mainly consider the historical same period gas transmission amount factor, select the historical same period gas transmission amount value of the previous two years, aggregate by year, take the average value, and refer to the X1 arithmetic sequence data to obtain the annual gas transmission prediction value, the data is calculated by recursion, and the X3 sequence is obtained;

[0072] The mathematical expression is:

[0073] Calculate the average gas amount of 3 days in 2019 and 2020 respectively;

[0074]

[0075] Use the 2019 and 2020 data to calculate the gas transmission amount in 2021;

[0076]

[0077] Finally, according to the function relationship between the influencing factors X and the predicted gas amount f(x) by Matlab data modeling software, the function fitting relationship of a certain user is as follows, that is, the prediction model is:

[0078] f(X1, X2, X3) = -0.002 + 0.1875*X1 + 0.8346*X2 + 0.0061*X3.

[0079] In the formula: X1 is the gas transmission amount under the consideration of historical same period temperature under the consideration of temperature factor; X2 is the average gas transmission amount under the consideration of recent N days gas transmission amount; X3 is the gas transmission amount under the consideration of historical same period gas transmission amount.

[0080] Prediction model index: R 2 The determination coefficient is 0.9117, which means that the performance of the final prediction model is excellent. (R 2 The value range is [0, 1], 0 represents the worst, and 1 represents the best. The above process is the modeling process of building a full-day prediction model for the example user.

[0081] Step S03: using the pipeline gas amount prediction model to predict the future scheduled gas amount of each user, and adjusting the real-time gas supply amount according to the scheduled gas amount prediction result.

[0082] In the present application, the method for predicting the future scheduled period gas consumption of each user by using the pipeline gas consumption prediction model comprises: predicting the total gas consumption of each user in the future scheduled number of hours by using the pipeline gas consumption prediction model; determining the hourly gas consumption proportion in the scheduled number of hours according to historical gas consumption data; and determining the future scheduled period gas consumption according to the hourly gas consumption proportion and the total gas consumption.

[0083] In the embodiments of the present application, the future scheduled period gas consumption can be the hourly gas consumption in 0-24 hours of the whole day or a certain time period (scheduled number of hours).

[0084] If the scheduled number of hours is 24 hours (for example, from 6 a.m. to 6 a.m. the next day), the total gas consumption of a certain user in the future 24 hours is first predicted by using the pipeline gas consumption prediction model. At the same time, the proportion of the hourly gas consumption in the total gas consumption in 24 hours also needs to be determined.

[0085] After the total gas consumption in the scheduled number of hours (i.e., the total gas consumption in 24 hours) and the proportion of the hourly gas consumption in the total gas consumption in 24 hours are determined, the specific hourly gas consumption can be obtained by multiplying the two.

[0086] In the present application, the method for determining the hourly gas consumption proportion in the scheduled number of hours according to historical gas consumption data comprises: finding the historical same-period gas consumption proportion data corresponding to the hour for which the gas consumption proportion is to be determined; determining the average value of the gas consumption proportions corresponding to the same hour in the previous scheduled different number of days of the date on which the historical same-period gas consumption proportion data is located; comparing the average value of each of the proportions obtained with the corresponding historical same-period gas consumption proportion to determine the correlation coefficient therebetween, and the scheduled different number of days corresponding to the maximum value of all the correlation coefficients is the optimal sampling period; and the hourly gas consumption proportion in the scheduled number of hours is the average value of the gas consumption proportions corresponding to the same hour in the previous optimal sampling period of days of the date on which each hour is located.

[0087] In the embodiments of the present application, to predict the hourly gas consumption proportion corresponding to each hour in 24 hours, the historical same-period gas consumption proportion data corresponding to the 24 hours needs to be found first. For example, if the future scheduled number of hours is from 6 a.m. on May 1, 2021 to 6 a.m. the next day, the historical same-period gas consumption proportion data corresponding thereto is the hourly gas consumption proportion between 6 a.m. on May 1, 2020 and 6 a.m. the next day.

[0088] The range for the number of scheduled days is 3 to 30 days. To predict the hourly gas usage percentage between 6:00 AM on May 1, 2021, and 6:00 AM the following day, we first need to determine the average gas usage percentage for all scheduled days preceding the date of the historical gas usage percentage data for the same period, corresponding to the hour of the to-be-determined gas usage percentage.

[0089] For example, if the reservations are made for different days, namely 3, 4, and 5; to determine the gas usage percentage in the future reservation hours, that is, the hour between 1:00 and 2:00 between 6:00 a.m. on May 1, 2021 and 6:00 a.m. the next day, it is necessary to find the date of the historical gas usage percentage data for the same period, that is, May 1, 2020, and the average of all gas usage percentages corresponding to 1:00 and 2:00 in the previous 3, 4, and 5 days.

[0090] Specifically, if the gas consumption ratios between 1:00 and 2:00 on April 26 to 29, 2020 are a, b, c, d, and e respectively, then the average ratio of the same hour from 1:00 to 2:00 within 3 days among the different days scheduled before May 1, 2020 is A = (c+d+e) / 3, the average ratio for 4 days is B = (b+c+d+e) / 4, and the average ratio for 5 days is C = (a+b+c+d+e) / 5.

[0091] Compare the obtained average values ​​of each proportion, i.e. A, B, and C, with the gas consumption proportion in the same period in history, i.e. the gas consumption proportion within one hour from 1:00 to 2:00 on May 1, 2020, and determine the correlation coefficient between each average value of the proportion, i.e. A, B, and C, and the gas consumption proportion in the same period in history. Find the average value of the proportion corresponding to the maximum value of all correlation coefficients. If it is A, then the predetermined different days corresponding to A, i.e. 3 days, is the optimal sampling period.

[0092] The gas usage percentage for each hour between 6:00 AM on May 1, 2021, and 6:00 AM the following day is the average of the gas usage percentages for the same time period over the three days prior to May 1, 2021. Specifically, the gas usage percentage between 1:00 AM and 2:00 AM on May 1, 2021, is calculated as (the gas usage percentage between 1:00 AM and 2:00 AM on April 28, 2021, + the gas usage percentage between 1:00 AM and 2:00 AM on April 29, 2021, + the gas usage percentage between 1:00 AM and 2:00 AM on April 30, 2021,) / 3.

[0093] By using historical data from the same period, we can calculate the proportion of hourly gas transmission volume in the total gas transmission volume of the day, aggregate it by the same hour, and calculate the average proportion of hourly gas transmission volume. The sampling period (predetermined for different days) is 3-30 days. We use cyclic calculation and compare it with the proportion of the next day to determine the optimal period.

[0094] After the total gas consumption and the proportion of gas consumption per hour are determined, the specific gas consumption per hour can be obtained by multiplying the two. The hourly gas consumption results predicted by the prediction model are compared with the corresponding actual gas consumption, and a fitting graph of the predicted gas consumption curve and the actual gas consumption curve is drawn. The results are as follows: Figure 2 As shown, from Figure 2 It can be seen that the prediction results of the prediction model are more accurate than the actual gas consumption, and the overall good rate reaches 75.79%.

[0095] Step S04: Determine whether there is a deviation between the gas consumption prediction result of the predetermined period and the actual gas consumption of the predetermined period. If so, determine the deviation index of the prediction result of the pipeline gas volume prediction model, correct the prediction result of the pipeline gas volume prediction model according to the deviation index, and adjust the real-time gas supply according to the corrected prediction result.

[0096] In the present invention, the method for determining the deviation index of the prediction result of the pipeline gas volume prediction model includes: determining the deviation index using formula (1);

[0097]

[0098] Where: F 预 To predict the remaining full-day weather, F 实 is the actual remaining full-day energy, The total gas volume is predicted for each hour excluding the hours that have occurred.

[0099] In the present invention, the method for correcting the prediction result of the pipeline gas volume prediction model according to the deviation index includes: correcting the prediction result of the pipeline gas volume prediction model according to the deviation index using formula (2);

[0100] SP 目标 =(a+β1×P+β2×K)×SP 预测 (2);

[0101] Where: SP 目标 is the corrected prediction result, a, β1, β2 are correction coefficients, P = P1 / P2, where P1 is the regulating valve inlet pressure, P2 is the regulating valve outlet pressure, SP 预测 This is the prediction result of the pipeline gas volume prediction model.

[0102] In the embodiment of the present application, the prediction model is based on a big data algorithm, that is, the more data, the smaller the deviation between the prediction result of the model and the actual gas volume, and in the initial prediction, the prediction result may be deviated due to the less historical parameter data involved; the self-learning optimization of the model is to obtain a deviation index K by comparing the initial prediction result with the actual occurred gas volume, and the next prediction result is corrected according to the deviation, so as to achieve the minimum deviation of the prediction result.

[0103] The deviation coefficient K is determined by formula (1), K is the actual remaining predicted total volume / predicted remaining total volume; the method for judging whether there is a deviation between the predicted gas volume in the predetermined period and the actual gas volume in the predetermined period is as follows: if the gas supply volume deviation degree K=1, the predicted gas volume is equal to the actual gas volume, which is no deviation, K>1, the actual gas volume is lower, and K<1, the actual gas volume is higher, therefore, K>1 or K<1 is a deviation.

[0104] The K value is calculated every 20 minutes, the obtained K value is substituted into formula (2) to calculate the corrected gas volume prediction result, and the flow set value of the PID control of the adjusting valve is corrected according to the corrected gas volume prediction result. In the actual operation of the long-distance pipeline device, it is required that the pipeline gas supply capacity is in a relatively balanced state, and the gas supply volume is adjusted through the correction method, so that the optimal balanced state of the pipeline gas supply capacity can be achieved.

[0105] In the correction of the prediction result, the real-time dynamic correction model of the gas volume is established for correction, the correction model adopts an error back propagation algorithm, which is a neural network algorithm, the prediction result is corrected through the deviation K between the predicted gas volume calculated by formula (1) and the actual occurred gas volume, that is, formula (2), the basic forward calculation mode of the neural network algorithm is as follows: sample-input layer-computation layer-output result, the correction coefficients of the input layer and the computation layer of the correction model are re-adjusted through the deviation, which is an error back correction process, until the output result is reasonable compared with the actual deviation, and then the back correction is stopped.

[0106] In the present application, it also includes: every interval of a predetermined time, newly generated user influencing factor data and corresponding actual gas volume data are input into the pipeline gas volume prediction model to train the model.

[0107] In the embodiment of the present application, the gas volume is constantly changing, and there are many uncertainties in the influencing factors, therefore, after the model is used for a period of time, the newly generated data is input into the model to optimize and train the model, so that the model accuracy can be further improved.

[0108] In the application, the method for regulating the real-time gas supply amount comprises: in the case of insufficient natural gas, uninterruptedly supplying gas to the first-level users among all users according to the gas consumption in the corresponding predicted result, and intermittently supplying gas to the second-level users; if the gas consumption of a user in the future predetermined period exceeds a predetermined value in the predicted result, the pre-stored natural gas in the gas pipeline is consumed in the future predetermined period of the user.

[0109] In the embodiment of the application, because the number of users of the long-distance pipeline is large, the users are divided into different types according to the gas consumption characteristics, such as CNG users (purely vehicle-mounted CNG, CNG mother station, etc.), livelihood users (i.e. residential gas, such as heating and cooking, etc.), commercial and enterprise users, etc., the gas consumption rules (i.e. the time period of the peak and valley of gas consumption, and the corresponding gas consumption, etc.) of different types of users are different, and each user needs to be analyzed separately; therefore, in order to ensure the accuracy of the prediction model, when the model is established, the gas consumption data of a single user in recent years (such as five years) is used as a sample, the error back propagation algorithm is used to analyze the all-day gas peak and valley rules, and a gas supply model with hourly proportion (gas consumption proportion) distribution is established by using a cyclic calculation method.

[0110] The types of natural gas users include oilfield production, livelihood gas, and key enterprises, etc., after the gas consumption of each user is predicted by the prediction model, when the gas supply is controlled, the key users and non-key users that need to maintain the gas supply amount are considered, the first-level users are the key users that need to ensure the gas supply amount, such as livelihood gas and oilfield production, and the second-level users are the users that can be intermittently supplied with gas, such as enterprise users. At the same time, when it is predicted that the gas consumption of a user in the next period increases, the pre-stored gas in the pipeline can be provided in advance, so as to prevent the situation that the gas supply flow cannot meet the requirements at the same time due to the simultaneous gas consumption peak of multiple users.

[0111] According to the prediction result, the gas consumption proportions of oilfield production, livelihood gas, and key enterprises can be determined, the gas supply flow of each period of the whole day is set and autonomously adjusted according to the gas consumption proportions, the second-level users among the interruptible downstream users are supplied with gas during the peak of livelihood gas, and intermittent gas supply is implemented, so as to ensure the gas supply amount of livelihood gas; during the valley of livelihood gas, other users are supplied with gas in sufficient amount, and the purpose of "peak shaving and valley filling" is achieved during the special period in winter. The autonomous adjustment refers to automatically adjusting the set value of the regulating valve of each gas supply node of the first station and the last station of the long-distance pipeline through the industrial control system according to the prediction result of the model, automatically setting the parameters according to the prediction result of each period, and automatically executing the gas supply and distribution through the automatically adjusted valve according to the pre-set PID adjustment mode, so as to reduce the labor intensity, improve the production efficiency, and achieve the purpose of improving the control accuracy.

[0112] It can be understood that the above-mentioned various method embodiments mentioned in the present invention can be combined with each other to form combined embodiments without violating the principle logic. Due to space limitations, the present invention will not elaborate on them.

[0113] The execution subject of the method for regulating the supply and demand relationship of a long-distance natural gas pipeline may be a device for regulating the supply and demand relationship of a long-distance natural gas pipeline. For example, the method for regulating the supply and demand relationship of a long-distance natural gas pipeline may be executed by a terminal device, a server, or other processing device, wherein the terminal device may be a user equipment (UE), a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, a personal digital assistant (PDA), a handheld device, a computing device, an in-vehicle device, a wearable device, etc. In some possible implementations, the method for regulating the supply and demand relationship of a long-distance natural gas pipeline may be implemented by a processor invoking computer-readable instructions stored in a memory.

[0114] Those skilled in the art will understand that in the above-mentioned method of the specific implementation method, the writing order of each step does not mean a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of each step should be determined by its function and possible internal logic.

[0115] In the present invention, a device for regulating the supply and demand relationship of a long-distance natural gas pipeline includes: an acquisition unit for acquiring historical influencing factor data and corresponding gas consumption data related to gas consumption prediction; a model establishment unit for establishing a multivariate parameter mathematical model between the historical influencing factor data and the gas consumption data based on the historical influencing factor data and the gas consumption data, wherein the multivariate parameter mathematical model is a pipeline gas consumption prediction model; a prediction unit for predicting the gas consumption of each user in a future scheduled period of time using the pipeline gas consumption prediction model, and regulating the real-time gas supply according to the predicted gas consumption result of the scheduled period; a correction unit for determining whether there is a deviation between the predicted gas consumption result of the scheduled period and the actual gas consumption in the scheduled period, and if so, determining a deviation index of the predicted result of the pipeline gas consumption prediction model, correcting the predicted result of the pipeline gas consumption prediction model according to the deviation index, and regulating the real-time gas supply according to the corrected predicted result.

[0116] In some embodiments, the functions or modules and units included in the device provided by the embodiment of the present invention can be used to execute the method described in the above method embodiment. Its specific implementation can refer to the description of the above method embodiment. For the sake of brevity, it will not be repeated here.

[0117] The application is based on the TRIZ pre-acting concept, and uses data dimension reduction analysis, model self-learning optimization and other key technologies to calculate the mathematical relationship between gas consumption and influencing factors such as gas temperature and historical data, and to establish a user gas consumption prediction model by comprehensively analyzing pipeline operation big data. The actual gas consumption data is used to verify the predicted gas consumption, the deviation index of the predicted value and the actual value of the prediction model is determined, the prediction result is corrected according to the deviation index, and a high-precision prediction model is generated, so that the gas consumption control realizes model intelligence, the pipeline supply and demand can be matched, and the import and export balance is achieved.

[0118] The application solves the problem that long-distance gas transmission in the gas transmission process of the natural gas long-distance pipeline causes a lag between supply and demand, and the supply and demand of the long-distance pipeline network are not matched, the gas consumption peak of the downstream user is predicted in advance, and gas is supplied in advance to ensure that the increased gas quantity reaches on time during the gas consumption peak of the user. According to the different upstream gas supply capacities of the natural gas long-distance pipeline network, the gas quantity distribution and regulation of various users in the pipeline network are automatically adjusted, and then the stable supply and demand of the long-distance pipeline network is ensured, the control precision of the key parameters of the gas supply pipeline network is improved, the self-supply of different types of gas users is realized, and the frequency of pressure fluctuation of the pipeline network is reduced. The industrial control system automatically controls the global regulation of the supply and demand relationship of the pipeline network, ensures reliable gas supply of the downstream user, and improves the control precision and work efficiency.

[0119] The application is tested and put into use in the natural gas long-distance pipeline, and after put into use, the control system can automatically distribute gas quantity to users according to the sequence of livelihood, key industry and commercial users, comprehensively consider factors such as import and export balance of the long-distance pipeline, pipeline inventory and gas supply weight, and intelligently distribute gas quantity according to the demand of users, and meanwhile, according to the different gas consumption rules of various users, the gas supply flow of each time period in a day is automatically set and adjusted, the control precision of hourly gas quantity can reach 99.8%, the operation frequency of post staff is reduced by 97%, the labor cost is saved by 1.45 million yuan per year, the gas supply quantity is accurate, the operation adaptability is strong, the reliable gas consumption of the downstream user is ensured, and good social benefits are obtained.

[0120] The above has described the embodiments of the application, the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes are obvious to those skilled in the art without departing from the scope and spirit of the described embodiments. The selection of the terms used herein is intended to best explain the principles, practical applications or technical improvements in the market of the embodiments, or to enable other ordinary skilled persons in the art to understand the embodiments disclosed herein.

Claims

1. A method for regulating the supply and demand relationship of a long-distance natural gas pipeline, characterized in that: include: Obtain historical influencing factor data related to gas consumption forecast and corresponding gas consumption data; Establishing a multivariate parameter mathematical model between the historical influencing factor data and the gas consumption data, wherein the multivariate parameter mathematical model is a pipeline gas consumption prediction model; The pipeline gas volume prediction model is used to predict the gas consumption of each user in a predetermined period in the future, and the real-time gas supply volume is adjusted according to the gas consumption prediction result in the predetermined period; Determine whether there is a deviation between the gas consumption prediction result of the predetermined period and the actual gas consumption of the predetermined period. If so, determine the deviation index of the prediction result of the pipeline gas volume prediction model, correct the prediction result of the pipeline gas volume prediction model according to the deviation index, and adjust the real-time gas supply according to the corrected prediction result.

2. The method for regulating supply and demand relationship of a long-distance natural gas pipeline according to claim 1, characterized in that: Before obtaining the historical influencing factor data related to the gas consumption forecast, the historical influencing factor data related to the gas consumption forecast is determined, and the method includes: Obtain historical pipeline operating parameter data, perform data dimensionality reduction analysis on the pipeline operating parameter data, and obtain historical influencing factor data related to gas consumption prediction.

3. The method for regulating supply and demand relationship of a long-distance natural gas pipeline according to claim 1, characterized in that: The influencing factor data at least includes: The flow rate of each user, the hourly gas volume of each user, the cumulative gas volume of each user throughout the day, the cumulative gas volume of each user per year, the natural gas flow rate in the pipeline, the pipeline inlet temperature, the pipeline outlet temperature, the pipeline temperature change, the pipeline inlet pressure, the pipeline outlet pressure, the pipeline inlet and outlet pressure difference, the pipeline pressure change, the average ambient temperature, the maximum temperature and the minimum temperature.

4. The method for regulating supply and demand relationship of a long-distance natural gas pipeline according to claim 1, characterized in that: The method for predicting the gas consumption of each user in a predetermined future period using the pipeline gas consumption prediction model includes: The pipeline gas volume prediction model is used to predict the total gas consumption of each user in the next predetermined number of hours; Determine the gas usage percentage for each hour within the predetermined number of hours based on historical gas usage data; The gas consumption in the future scheduled period is determined based on the gas consumption proportion and the total gas consumption.

5. The method for regulating supply and demand relationship of a long-distance natural gas pipeline according to claim 4, characterized in that: The method for determining the gas usage percentage of each hour within the predetermined number of hours based on historical gas usage data includes: Find the historical gas consumption rate data for the hour in which the gas consumption rate is to be determined; Determine the average value of the gas consumption ratios corresponding to the same hour as the gas consumption ratio to be determined in predetermined different days before the date of the historical gas consumption ratio data for the same period; The average value of each of the proportions obtained is compared with the corresponding historical gas consumption proportion in the same period to determine the correlation coefficient between the two, and the predetermined different days corresponding to the maximum value of all the correlation coefficients are the optimal sampling period; The gas usage ratio of each hour in the predetermined number of hours is: the average gas usage ratio of the same hour corresponding to each hour in the previous optimal sampling period days of the date of each hour.

6. The method for regulating supply and demand relationship of a long-distance natural gas pipeline according to claim 1, characterized in that: The method for determining the deviation index of the prediction result of the pipeline gas volume prediction model includes: Determine the deviation index using formula (1); Where: F 预 To predict the remaining full-day weather, F 实 is the actual remaining full-day energy, The total gas volume is predicted for each hour excluding the hours that have occurred.

7. The method for regulating supply and demand relationship of a long-distance natural gas pipeline according to claim 6, characterized in that: The method for correcting the prediction result of the pipeline gas volume prediction model according to the deviation index includes: According to the deviation index, the prediction result of the pipeline gas volume prediction model is corrected using formula (2); SP 目标 =(a+β1×P+β2×K))×SP 预测 (2); Where: SP 目标 is the corrected prediction result, a, β1, β2 are correction coefficients, P = P1 / P2, where P1 is the regulating valve inlet pressure, P2 is the regulating valve outlet pressure, SP 预测 This is the prediction result of the pipeline gas volume prediction model.

8. The method for regulating supply and demand relationship of a long-distance natural gas pipeline according to claim 1, characterized in that: Also includes: At predetermined intervals, newly generated user influencing factor data and corresponding actual gas usage data are input into the pipeline gas volume prediction model to train the model.

9. The method for regulating supply and demand relationship of a long-distance natural gas pipeline according to any one of claims 1 to 8, characterized in that: The method for regulating the real-time gas supply volume includes: In the event of natural gas shortage, all first-level users will be supplied with uninterrupted gas according to their corresponding predicted gas consumption, and second-level users will be supplied with intermittent gas; If the gas consumption in a future scheduled period in the prediction result of a certain user exceeds a predetermined value, the natural gas pre-stored in the gas pipeline will be consumed during the future scheduled period of the user.

10. A natural gas long-distance pipeline supply and demand relationship control device, characterized in that: include: An acquisition unit, configured to acquire historical influencing factor data related to gas consumption prediction and corresponding gas consumption data; A model building unit, configured to build a multivariate parameter mathematical model between the historical influencing factor data and the gas consumption data, wherein the multivariate parameter mathematical model is a pipeline gas consumption prediction model; A prediction unit, configured to use the pipeline gas volume prediction model to predict the gas consumption of each user in a predetermined period in the future, and to regulate the real-time gas supply according to the predicted gas consumption results in the predetermined period; The correction unit is used to determine whether there is a deviation between the gas consumption prediction result of the predetermined time period and the actual gas consumption of the predetermined time period. If so, the deviation index of the prediction result of the pipeline gas volume prediction model is determined, the prediction result of the pipeline gas volume prediction model is corrected according to the deviation index, and the real-time gas supply is regulated according to the corrected prediction result.

Images