Method and device for predicting monthly gas consumption sales volume of urban gas in cold and warm winter

Abstract

The invention relates to a cold and warm winter urban gas monthly gas consumption sales volume prediction method and device, and belongs to the technical field of natural gas sales volume prediction.The prediction method comprises the steps that the monthly gas consumption sales volume of a to-be-measured month in the normal temperature winter is obtained based on the annual total natural gas demand and the daily unevenness coefficient in the normal temperature winter; on the basis of the multiple predicted temperatures of the cold winter or the warm winter and the relation curve of the daily unevenness coefficient and the temperature, the monthly gas consumption sales corresponding to the multiple predicted temperatures of the cold winter or the warm winter is obtained; and calculating the increasing proportion of the monthly gas consumption sales of cold winter or the decreasing proportion of the monthly gas consumption sales of warm winter based on the monthly gas consumption sales of the monthly to-be-measured month at normal temperature in winter and the monthly gas consumption sales corresponding to a plurality of predicted temperatures of cold winter or warm winter, and establishing a database of the influence of temperature change on the monthly gas consumption sales. According to the method or device provided by the invention, the dependence on prediction experience is reduced, and the accuracy of gas consumption sales data prediction is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of natural gas sales forecasting, and in particular relates to a method for forecasting monthly urban gas sales in cold and warm winters. Background Art

[0002] As China's gas consumption continues to rise, gas facilities in towns and villages continue to improve, my country's natural gas industry grows rapidly, and natural gas consumption continues to increase. Heating gas is the main component of winter gas consumption of urban gas companies in northern regions, and temperature changes have a significant impact on daily load changes. With the improvement of people's material life, the demand for winter heating in southern regions has also increased, and many residents use natural gas for heating and domestic hot water.

[0003] At present, city gas companies rely on the operating experience of operation and dispatching personnel to predict demand in the gas supply area, and estimate natural gas demand based on the monthly gas consumption of the previous year and the market development of this year. However, this forecasting method does not take into account the changes in demand caused by the rapid changes in temperature in the gas supply area. Therefore, it is necessary to select a suitable forecasting method based on the analysis of the mechanism of the impact of temperature on heating gas load, reduce dependence on forecasting experience, and achieve the goal of improving forecasting accuracy.

[0004] At present, the monthly urban gas consumption prediction method in the related art is only based on the prediction of historical rules. When the temperature in the heating season changes significantly (i.e., warm winter or cold winter), the monthly urban gas consumption in this case cannot be obtained.

[0005] Based on this, the present invention provides a method and device for predicting monthly city gas consumption and sales volume in cold and warm winters. Summary of the invention

[0006] In order to at least partially solve the technical problem of high reliance on prediction experience in the related art, an embodiment of the present invention provides a method and device for predicting monthly city gas consumption and sales volume in cold and warm winters.

[0007] In a first aspect, an embodiment of the present invention provides a method for predicting monthly city gas consumption and sales volume in cold and warm winters. The prediction method comprises:

[0008] Based on the annual total natural gas demand and the daily unevenness coefficient in winter with normal temperature, the monthly gas sales volume in the test month in winter with normal temperature is obtained, wherein the daily unevenness coefficient is defined as daily gas consumption / average daily gas consumption, and the average daily gas consumption is defined as annual gas sales volume / 365 days;

[0009] Based on multiple predicted temperatures of cold winter or warm winter and the relationship curve between daily degree unevenness coefficient and temperature, the monthly gas sales volume corresponding to multiple predicted temperatures of cold winter or warm winter is obtained;

[0010] Based on the monthly gas sales of the test month in a normal winter and the monthly gas sales corresponding to multiple predicted temperatures in a cold winter or a warm winter, the increase ratio of the monthly gas sales in a cold winter or the decrease ratio of the monthly gas sales in a warm winter is calculated, and a database of the impact of temperature changes on monthly gas sales is established;

[0011] Based on the temperature of the month to be predicted, the database of the impact of temperature changes on monthly gas sales, and the non-final predicted monthly gas sales of the predicted month, the final predicted monthly gas sales for cold and warm winters are calculated.

[0012] In some embodiments, the total annual natural gas demand is predicted by a grey GM (1, 1) model.

[0013] In some embodiments, the step of obtaining the monthly gas sales volume of the test month in winter with normal temperature based on the annual total natural gas demand and the daily unevenness coefficient in winter with normal temperature includes:

[0014] Calculate the product of the total annual natural gas demand and the daily uneven coefficient in winter with normal temperature to obtain the daily gas sales volume in winter under normal conditions;

[0015] The daily gas sales corresponding to different dates in the test month in winter with normal temperature are added up to obtain the monthly gas sales of the test month in winter with normal temperature.

[0016] In some embodiments, the determination of the plurality of predicted temperatures comprises:

[0017] Based on the historical average daily temperature in winter, a plurality of predicted temperatures for a cold winter is determined by a decreasing gradient method, or a plurality of predicted temperatures for a warm winter is determined by an increasing gradient method.

[0018] In some embodiments, obtaining the relationship curve between the daily non-uniformity coefficient and the temperature includes:

[0019] Based on the historical data of urban temperature and historical natural gas sales data, the benchmark temperature is screened out;

[0020] According to the historical sales data of natural gas and the benchmark temperature, the relationship matrix between the daily degree unevenness coefficient and the benchmark temperature is constructed;

[0021] Based on the relationship matrix between daily degree unevenness coefficient and reference temperature, the abnormal points of the relationship between daily degree unevenness coefficient and reference temperature are eliminated to obtain the data points of daily degree unevenness coefficient and reference temperature;

[0022] Based on the daily degree unevenness coefficient and the reference temperature data points, a piecewise linear fitting curve of the relationship between the daily degree unevenness coefficient and the temperature number is performed according to the reference temperature interval.

[0023] In some embodiments, the temperature history data includes the maximum temperature, the minimum temperature and the average temperature of each day, and the natural gas historical sales data includes the natural gas sales of each day.

[0024] In some embodiments, in the step of filtering out the reference temperature based on the historical city temperature data and the historical natural gas sales data, the specific operation of filtering is as follows:

[0025] Compare the correlation coefficients between the highest temperature and natural gas sales, the lowest temperature and natural gas sales, and the average temperature and natural gas sales, and select the temperature with the largest correlation coefficient as the benchmark temperature.

[0026] In some embodiments, obtaining monthly gas sales corresponding to multiple predicted temperatures of a cold winter or a warm winter based on multiple predicted temperatures of a cold winter or a warm winter and a relationship curve between a daily degree unevenness coefficient and a temperature includes:

[0027] Substituting the specific predicted temperature of the cold winter or warm winter into the relationship curve between the daily degree unevenness coefficient and the temperature, the daily degree unevenness coefficient under the specific predicted temperature of the cold winter or warm winter is obtained;

[0028] Calculate the product of the annual total natural gas demand and the daily unevenness coefficient under the specific predicted temperature of cold winter or warm winter to obtain the daily gas sales volume under the specific predicted temperature of cold winter or warm winter;

[0029] The daily gas sales volume under the specific predicted temperature of cold winter or warm winter corresponding to different dates in a month is added up to obtain the monthly gas sales volume corresponding to the specific predicted temperature of cold winter or warm winter;

[0030] The calculation of daily unevenness coefficient, daily gas sales volume and monthly gas sales volume under different predicted temperatures is repeated to obtain monthly gas sales volume corresponding to multiple predicted temperatures of cold winter or warm winter.

[0031] In some embodiments, the predicted monthly gas sales volume is not the final predicted amount, which is the monthly gas sales volume of the month to be tested in winter with normal temperature or the monthly gas sales volume of the month to be tested predicted based on historical monthly gas sales data.

[0032] In a second aspect, an embodiment of the present invention provides a device for predicting monthly city gas consumption and sales volume in cold and warm winters. The prediction device comprises: a normal temperature winter prediction unit, a cold and warm winter prediction unit, a database construction unit, and a calculation unit;

[0033] The normal temperature winter prediction unit is used to obtain the monthly gas sales volume of the test month in the normal temperature winter based on the annual total natural gas demand and the daily uneven coefficient of the normal temperature winter, wherein the daily uneven coefficient is defined as the daily gas consumption / average daily gas consumption, and the average daily gas consumption is defined as the annual gas sales volume / 365 days;

[0034] The cold and warm winter prediction unit is used to obtain monthly gas sales corresponding to multiple predicted temperatures of cold winter or warm winter based on multiple predicted temperatures of cold winter or warm winter and a relationship curve between daily degree unevenness coefficient and temperature;

[0035] The database construction unit is used to calculate the increase ratio of monthly gas sales in cold winter or the decrease ratio of monthly gas sales in warm winter based on the monthly gas sales in the test month in normal temperature winter and the monthly gas sales corresponding to multiple predicted temperatures in cold winter or warm winter, and to build a database of the impact of temperature changes on monthly gas sales;

[0036] The calculation unit is used to calculate the final predicted monthly gas sales volume for cold and warm winters based on the temperature of the month to be predicted, the database of the impact of temperature changes on monthly gas sales volume, and the non-final predicted monthly gas sales volume of the predicted month.

[0037] In a third aspect, an embodiment of the present invention provides an electronic device. The electronic device comprises: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus;

[0038] A memory for storing programs and / or instructions;

[0039] A processor, configured to execute the program and / or instructions;

[0040] Wherein, when the processor executes the program and / or instruction, it implements the above-mentioned method for predicting monthly gas consumption and sales volume of urban gas in cold and warm winters.

[0041] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method for predicting monthly gas consumption and sales volume of urban gas in cold and warm winters as described above is implemented.

[0042] The above technical solution provided by the embodiment of the present invention has at least some or all of the following advantages:

[0043] Through the pre-constructed database of the ratio of temperature changes in cold and warm winters to gas sales changes, the monthly gas sales data of the target month predicted by other models can be corrected, reducing dependence on forecasting experience and improving the accuracy of gas sales data forecasts.

[0044] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0046] Figure 1 A flowchart of a method for predicting monthly city gas consumption and sales volume in cold and warm winters according to an embodiment of the present invention is shown;

[0047] Figure 2 It shows a structural framework diagram of a device for predicting monthly city gas consumption and sales volume in cold and warm winters according to an embodiment of the present invention;

[0048] Figure 3 A structural framework diagram of an electronic device according to an embodiment of the present invention is shown;

[0049] In the figure:

[0050] 100. Gas consumption and sales volume prediction device; 101. Normal temperature winter prediction unit; 102. Cold and warm winter prediction unit; 103. Database construction unit; 104. Calculation unit; 200. Electronic device; 201. Processor; 202. Communication interface; 203. Memory; 204. Communication bus. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] A first exemplary embodiment of the present invention provides a method for predicting monthly city gas consumption and sales volume in cold and warm winters.

[0053] Figure 1 A flowchart of a method for predicting monthly city gas consumption and sales volume in cold and warm winters provided by an embodiment of the present invention is schematically shown.

[0054] Reference Figure 1 As shown, the embodiment of the present invention provides a method for predicting monthly city gas consumption and sales volume in cold and warm winters based on temperature changes, including:

[0055] Based on the annual total natural gas demand and the daily unevenness coefficient in winter with normal temperature, the monthly gas sales volume in the test month in winter with normal temperature is obtained, wherein the daily unevenness coefficient is defined as daily gas consumption / average daily gas consumption, and the average daily gas consumption is defined as annual gas sales volume / 365 days;

[0056] Based on multiple predicted temperatures of cold winter or warm winter and the relationship curve between daily degree unevenness coefficient and temperature, the monthly gas sales volume corresponding to multiple predicted temperatures of cold winter or warm winter is obtained;

[0057] Based on the monthly gas sales of the test month in a normal winter and the monthly gas sales corresponding to multiple predicted temperatures in a cold winter or a warm winter, the increase ratio of the monthly gas sales in a cold winter or the decrease ratio of the monthly gas sales in a warm winter is calculated, and a database of the impact of temperature changes on monthly gas sales is established;

[0058] Based on the temperature of the month to be predicted, the database of the impact of temperature changes on monthly gas sales, and the non-final predicted monthly gas sales of the predicted month, the final predicted monthly gas sales for cold and warm winters are calculated.

[0059] Specifically, the monthly gas sales volume of the test month in winter with normal temperature is obtained based on the annual total natural gas demand and the daily unevenness coefficient in winter with normal temperature, including:

[0060] The total annual natural gas demand is predicted by the grey GM (1, 1) model. The total annual natural gas demand may also be predicted by other existing prediction models and then by the calculation logic preset in the model;

[0061] Calculate the product of the total annual natural gas demand and the daily uneven coefficient in winter with normal temperature to get the daily gas sales volume in winter under normal conditions;

[0062] The daily gas sales corresponding to different dates in the test month in winter with normal temperature are added up to obtain the monthly gas sales of the test month in winter with normal temperature.

[0063] In the step of obtaining monthly gas sales corresponding to multiple predicted temperatures of cold winter or warm winter based on multiple predicted temperatures of cold winter or warm winter and relationship curves between daily degree unevenness coefficient and temperature, the multiple predicted temperatures include the following steps to determine:

[0064] Based on the historical average daily temperature in winter, multiple predicted temperatures for cold winters are determined using a decreasing gradient method, or multiple predicted temperatures for warm winters are determined using an increasing gradient method;

[0065] Specifically, in this embodiment, the average daily temperature T0 of the winter of the previous 5 years is taken, and the specific historical time limit can also be the previous 2 years, the previous 3 years, the previous 4 years, the previous 6 years, the previous 9 years, and the previous 10 years, and the historical years in the historical time limit can be adjacent or non-adjacent historical years, and the closer to the predicted year, the better. The specific historical time limit is not limited in the present invention;

[0066] Based on the historical winter daily average temperature T0:

[0067] In a cold winter, a plurality of predicted temperatures for the cold winter are determined by a gradient decreasing method, wherein the decreasing temperature corresponding to the gradient decreasing is 0.1°C, and the number of decreasing times is n times, then the plurality of predicted temperatures corresponding to the cold winter are T0-0.1, T0-0.2, ..., T0-0.n;

[0068] In the case of a warm winter, a plurality of predicted temperatures for the warm winter are determined by a gradient increasing method, wherein the increasing temperature corresponding to the gradient decreasing is 0.1°C, and the number of increasing times is n times, then the plurality of predicted temperatures corresponding to the warm winter are T0+0.1, T0+0.2, ..., T0+0.n;

[0069] The above-mentioned decreasing temperature or increasing temperature can also be 0.01°C or 0.02°C or 0.03°C or 0.05°C or 0.2°C or 0.5°C or 1°C or 1.5°C or 2°C. In the present invention, the decreasing temperature and the increasing temperature are not limited;

[0070] The above-mentioned decreasing or increasing times is 30 times, 50 times, 100 times, 200 times, 400 times or 500 times. In the present invention, there is no limitation on the decreasing or increasing times.

[0071] In the step of obtaining monthly gas sales corresponding to multiple predicted temperatures of cold winter or warm winter based on multiple predicted temperatures of cold winter or warm winter and a relationship curve between daily degree unevenness coefficient and temperature, the relationship curve between daily degree unevenness coefficient and temperature includes the following steps to obtain:

[0072] Based on the historical data of urban temperature and historical natural gas sales data, the benchmark temperature is screened out;

[0073] According to the historical sales data of natural gas and the benchmark temperature, the relationship matrix between the daily degree unevenness coefficient and the benchmark temperature is constructed;

[0074] Based on the relationship matrix between daily degree unevenness coefficient and reference temperature, the abnormal points of the relationship between daily degree unevenness coefficient and reference temperature are eliminated to obtain the data points of daily degree unevenness coefficient and reference temperature;

[0075] Based on the daily degree unevenness coefficient and the reference temperature data points, a piecewise linear fitting curve of the relationship between the daily degree unevenness coefficient and the temperature number is performed according to the reference temperature interval.

[0076] The city temperature history data includes the highest temperature, lowest temperature and average temperature of each day in the historical winter. For example, the highest temperature, lowest temperature and average temperature of each day in the previous five winters are taken. Specifically, if the gas sales volume in December of the winter of 2022 is predicted, the highest temperature, lowest temperature and average temperature of each day in December 2021, December 2020, December 2019, December 2018 and December 2017 are taken.

[0077] The natural gas historical sales data includes daily natural gas sales, for example, daily natural gas sales in the previous five winters. Specifically, to predict the gas sales in December of the winter of 2022, the corresponding natural gas sales (daily natural gas sales) in December 2021, December 2020, December 2019, December 2018 and December 2017 are taken;

[0078] In some embodiments, the maximum temperature, minimum temperature and average temperature of each day, and the selected historical time of the daily natural gas sales volume may also be the maximum temperature, minimum temperature and average temperature of each day in any one or more years before the year to be predicted, and the daily natural gas sales volume, which will not be described in detail in the present invention;

[0079] The base temperature is screened out based on the historical city temperature data and the historical natural gas sales data, specifically including:

[0080] Based on the data of daily maximum temperature, minimum temperature and average temperature, and daily natural gas sales volume, by constructing curves of the above four data (maximum temperature, minimum temperature, average temperature and daily natural gas sales volume) and time, the correlation coefficient between the maximum temperature and daily natural gas sales volume, the correlation coefficient between the minimum temperature and daily natural gas sales volume, and the correlation coefficient between the average temperature and daily natural gas sales volume are obtained. Then, the sizes of the three correlation coefficients are compared, and the temperature corresponding to the largest correlation coefficient is selected as the reference temperature. The temperature selected by this method can maximize the impact of temperature changes on natural gas sales volume and improve the accuracy of natural gas sales prediction.

[0081] After the above steps of screening out the benchmark temperature, combined with the daily natural gas sales data, a relationship matrix between the daily unevenness coefficient and the benchmark temperature is constructed, including:

[0082] If, in the process of obtaining the above-mentioned historical urban temperature data and historical natural gas sales data, the selected historical time for data selection is annual, the annual interval needs to be divided into a heating season (usually from November 15th to March 15th) and a non-heating season (from March 16th to November 14th). Correspondingly, the daily uneven coefficient is also divided into the heating season daily uneven coefficient and the non-heating season daily uneven coefficient. The heating season daily uneven coefficient is defined as the daily gas consumption divided by the average daily gas consumption in the heating season. The average daily gas consumption in the heating season is defined as the gas consumption in the heating season divided by the number of days in the heating season. The non-heating season daily uneven coefficient is defined as the daily gas consumption divided by the average daily gas consumption in the non-heating season. The average daily gas consumption in the non-heating season is defined as the gas consumption in the non-heating season divided by the number of days in the non-heating season. The obtained relationship matrix between the daily uneven coefficient and the reference temperature is the relationship matrix between the heating season daily uneven coefficient and the reference temperature, and the relationship matrix between the non-heating season daily uneven coefficient and the reference temperature;

[0083] If, in the process of obtaining the above-mentioned historical urban temperature data and historical natural gas sales data, the selected historical time for data is the data corresponding to winter, then no division is required.

[0084] After constructing the relationship matrix between the daily uneven coefficient and the reference temperature, combined with the relationship matrix between the daily uneven coefficient and the reference temperature, the abnormal points in the relationship between the daily uneven coefficient and the reference temperature are removed to obtain the data points of the daily uneven coefficient and the reference temperature. Among them, the abnormal points refer to the points where the natural gas sales are low or high due to external factors (such as gas pipeline maintenance, emergencies). The screening method for abnormal points is to use the one-class SVM algorithm for screening, and then remove them after screening.

[0085] Based on the data points of the daily uneven coefficient and the reference temperature, the relationship curve between the daily uneven coefficient and the temperature number is segmented and linearly fitted according to the interval of the reference temperature. Specifically, it includes:

[0086] The reference temperature is divided into multiple temperature intervals. The division method of the multiple temperature intervals can use the CART (Classification and Regression Tree) algorithm for division to obtain the dataset of the daily natural gas sales and the reference temperature corresponding to multiple temperature intervals. The multiple temperature intervals obtained after division are, for example, D1 interval: T'<t1; D2 interval: t1≤T'<t2; D3 interval: t2≤T'<t3... Dn interval: t n-1 ≤T'<t n where T' is the reference temperature, and t1, t2, t3... t n-1 and t n are the boundary temperature values of the temperature intervals.

[0087] According to the data sets of daily natural gas sales and benchmark temperatures in multiple temperature intervals obtained above, the daily degree unevenness coefficient is calculated according to the definition of the daily degree unevenness coefficient, and then the data sets of daily degree unevenness coefficients and benchmark temperatures in multiple temperature intervals are obtained;

[0088] According to the data sets of daily degree unevenness coefficient and reference temperature in multiple temperature intervals, the relationship curve between daily degree unevenness coefficient and temperature is fitted piecewise linearly. The curve relationship of multiple intervals obtained by fitting is as follows:

[0089]

[0090] In the above formula, y is the daily unevenness coefficient, T is the temperature, a1, a2, a3...a n is the slope of the straight line obtained by fitting the daily nonuniformity coefficient and temperature, b1, b2, b3...b n is the intercept of the straight line obtained by fitting the daily degree nonuniformity coefficient and temperature on the temperature axis. After fitting, a1, a2 and a3...a n , b1, b2 and b3...b n All are known numbers;

[0091] Similarly, if the historical data selection time is selected as annual in the process of obtaining the above-mentioned city temperature historical data and natural gas historical sales data, the annual interval needs to be divided into the heating season (generally November 15th to March 15th) and the non-heating season (March 16th to November 14th), and the temperature interval is divided into different time periods, and the fitting of the relationship curve between the daily degree unevenness coefficient and the temperature in the heating season and the fitting of the relationship curve between the daily degree unevenness coefficient and the temperature in the non-heating season are performed respectively;

[0092] If, in the process of obtaining the aforementioned city temperature historical data and natural gas sales historical data, the data corresponding to the historical time of winter is selected, no division is required.

[0093] Specifically, the obtaining of monthly gas sales corresponding to the multiple predicted temperatures of the cold winter or warm winter based on the multiple predicted temperatures of the cold winter or warm winter and the relationship curve between the daily degree unevenness coefficient and the temperature includes:

[0094] Substitute the above-determined specific predicted temperature of the cold winter or warm winter into the above-determined relationship curve between the daily degree unevenness coefficient and the temperature to obtain the daily degree unevenness coefficient at the specific predicted temperature of the cold winter or warm winter, such as:

[0095] Compare T0-0.1 with D1 interval, D2 interval...Dn interval. When T0-0.1 falls into a specific temperature interval (such as D1 interval), directly substitute T0-0.1 into the formula y=a1*T+b1 to obtain the daily unevenness coefficient corresponding to T0-0.1.

[0096] The total annual natural gas demand obtained by the above prediction is multiplied by the daily unevenness coefficient under the specific predicted temperature of the cold winter or warm winter obtained above, and the product obtained is the daily gas sales volume under the specific predicted temperature of the cold winter or warm winter;

[0097] The daily gas sales volume under the specific predicted temperature of cold winter or warm winter corresponding to different dates in the month is added up to obtain the monthly gas sales volume corresponding to the specific predicted temperature of cold winter or warm winter, such as:

[0098] The forecast month is December 2022. Calculate the daily gas sales volume at the specific forecast temperature on December 1, 2022 (assuming it is T0-0.1), the daily gas sales volume at the specific forecast temperature on December 2, 2022 (T0'-0.1, T0' is the average temperature of December 2 in the same historical period)... and the daily gas sales volume at the specific forecast temperature on December 31, 2022 (T0'-0.1, T0 n The daily gas sales volume under the predicted temperature of the cold winter body is obtained by adding up the daily gas sales volume obtained above. That is, when predicting the monthly sales volume, the temperature gradients of the predicted temperatures of each day corresponding to different dates in the month should be the same, i.e., the decreasing or increasing gradients should be used.

[0099] In certain embodiments, in order to further improve the accuracy of the predicted monthly natural gas sales, since the predicted month is December 31, 2022, which is a holiday, the dates for using the above calculation method are December 1, 2022 to December 30, 2022, and the daily gas sales on December 31, 2022 are predicted separately. The specific method is as follows:

[0100] Determine the holiday day unevenness coefficient of the day to be predicted;

[0101] Multiply the predicted daily average natural gas sales volume of the holiday to be predicted by the holiday daily unevenness coefficient, and the product is the daily gas sales volume of the day to be predicted.

[0102] Specifically, in this embodiment, December 31, 2022 is the New Year's Day holiday, and the calculation formula of the holiday day unevenness coefficient is as follows:

[0103]

[0104] Where A represents the holiday daily unevenness coefficient, k represents the number of years selected in the historical time limit, Ji represents the historical daily natural gas sales volume of the same period as the forecast date in the i-th year among the k-years, and Wi represents the average value of the historical natural gas sales volume of the preset number of days adjacent to the holiday of the i-th year among the k-years;

[0105] Specifically, the number of years selected for the historical time limit is the previous 5 years, J1 represents the daily natural gas sales on December 31, 2021, J2 represents the daily natural gas sales on December 31, 2020, J3 represents the daily natural gas sales on December 31, 2019, J4 represents the daily natural gas sales on December 31, 2018, J5 represents the daily natural gas sales on December 31, 2017, and W1 represents the adjacent preset days of the 2021 New Year's Day holiday (the adjacent days are set according to the specific holidays, the Spring Festival holiday takes the first 4 days of the Spring Festival holiday, the Spring Festival holiday and the last 3 days, a total of 14 days, other holidays take the 7 days before the holiday and the 7 days after the holiday, a total of 17 days) W1 represents the average value of historical natural gas sales (7 days before and 7 days after the New Year's Day holiday), W2 represents the average value of historical natural gas sales of the preset number of consecutive days during the 2020 New Year's Day holiday, W3 represents the average value of historical natural gas sales of the preset number of consecutive days during the 2019 New Year's Day holiday, W4 represents the average value of historical natural gas sales of the preset number of consecutive days during the 2018 New Year's Day holiday, and W5 represents the average value of historical natural gas sales of the preset number of consecutive days during the 2017 New Year's Day holiday. Substituting J1, J2, J3, J4, J5, W1, W3, W3, W4 and W5 into the above formula, the holiday daily unevenness coefficient on December 31, 2022 can be calculated;

[0106] The predicted daily average natural gas sales volume of the holiday to be predicted is multiplied by the daily uneven coefficient of the holiday, and the product obtained is the daily gas sales volume of the day to be predicted, including:

[0107] Calculate the predicted average daily natural gas sales volume for the holidays to be predicted. In this embodiment, the predicted average daily natural gas sales volume on December 31, 2022 is calculated by multiplying the total demand in the heating season by the daily uneven coefficient of the heating season, wherein the total demand in the heating season is obtained by the prediction and calculation of the gray GM (1, 1) model;

[0108] If the forecast day in the forecast month is during the Spring Festival holiday, the calculation formula for the holiday day unevenness coefficient is as follows:

[0109]

[0110] Among them, A represents the holiday daily non-uniformity coefficient, k represents the number of years selected in the historical time limit, Ji represents the historical daily natural gas sales volume in the same period as the forecast day in the i-th year in the k-th year, and Vi represents the historical daily average natural gas sales volume in the i-th heating season in the k-th year;

[0111] The predicted average daily natural gas sales volume for the holidays to be predicted during the Spring Festival holidays is calculated according to the predicted average daily natural gas sales volume to be tested in the heating season, that is, the total natural gas demand in the heating season is multiplied by the daily unevenness coefficient of the heating season. The total natural gas demand in the heating season can be obtained by prediction and calculation according to the gray GM (1, 1) model.

[0112] Predict the monthly gas sales under the temperature T0-0.2, that is, repeat the calculation of daily uneven coefficient, daily gas sales and monthly gas sales under T0-0.2;

[0113] …

[0114] Predict the monthly gas sales volume under the temperature T0-0.n, i.e. repeat the calculation of daily unevenness coefficient, daily gas sales volume and monthly gas sales volume under the specific predicted temperature;

[0115] Finally, the monthly gas sales corresponding to multiple predicted temperatures (T0-0.1, T0-0.2, ..., T0-0.n) of the cold winter are obtained;

[0116] The calculation of monthly gas sales corresponding to multiple predicted temperatures in a warm winter is the same as above.

[0117] Based on the monthly gas sales of the test month in winter with normal temperature (calculated by the annual total natural gas demand predicted by the gray GM (1, 1) model) and the monthly gas sales corresponding to the multiple predicted temperatures of the cold winter or warm winter, the increase ratio of the monthly gas sales in cold winter or the decrease ratio of the monthly gas sales in warm winter is calculated, and a database of the impact of temperature changes on monthly gas sales is established. In this embodiment, in order to briefly illustrate the database of the impact of temperature changes on monthly gas sales, a gradient decrease or a gradient increase of 1°C is used in determining the predicted temperature, and the number of decreases n is 3. The database of the impact of temperature changes on monthly gas sales is finally obtained as shown in Table 1. The normal temperature sales in the table correspond to the monthly gas sales of the test month in winter with normal temperature.

[0118] Table 1

[0119]

[0120] Specifically, the calculation of the final forecast amount of monthly gas sales for cold and warm winters based on the temperature of the month to be forecasted, the database of the impact of temperature changes on monthly gas sales, and the non-final forecast amount of monthly gas sales for the forecasted month includes:

[0121] The temperature of the month to be predicted is input into a database of the impact of temperature changes on monthly gas sales, and an increase or decrease ratio matching the temperature of the month to be predicted is obtained;

[0122] Multiply the non-final forecast amount of monthly gas sales in the forecast month by the increase or decrease ratio to obtain the increase or decrease in monthly gas sales in the forecast month;

[0123] The monthly gas sales volume of the forecast month is added or subtracted from the increase or decrease in the monthly gas sales volume of the forecast month to obtain the final forecast volume of the monthly gas sales volume of the cold and warm winter.

[0124] Among them, the non-final predicted monthly gas sales volume of the predicted month can be the monthly gas sales volume of the month to be tested in winter with normal temperature (calculated by the annual total natural gas demand predicted by the gray GM (1, 1) model); or the monthly gas sales volume of the month to be tested obtained by other prediction models, such as the monthly gas sales volume of the month to be tested predicted based on historical monthly gas sales data.

[0125] A second exemplary embodiment of the present invention provides a device for predicting monthly city gas consumption and sales volume in cold and warm winters.

[0126] Figure 2 A device for predicting monthly city gas consumption and sales volume in cold and warm winters provided by an embodiment of the present invention is schematically shown.

[0127] Reference Figure 2 As shown, the cold and warm winter city gas monthly gas consumption and sales volume prediction device 100 provided by the embodiment of the present invention includes: a normal temperature winter prediction unit 101, a cold and warm winter prediction unit 102, a database construction unit 103 and a calculation unit 104;

[0128] The normal temperature winter prediction unit 101 is used to obtain the monthly gas sales volume of the test month in the normal temperature winter based on the annual total natural gas demand and the daily uneven coefficient in the normal winter, wherein the daily uneven coefficient is defined as the daily gas consumption / average daily gas consumption, and the average daily gas consumption is defined as the annual gas sales volume / 365 days;

[0129] The cold and warm winter prediction unit 102 is used to obtain monthly gas sales corresponding to multiple predicted temperatures of cold winter or warm winter based on multiple predicted temperatures of cold winter or warm winter and a relationship curve between daily degree unevenness coefficient and temperature;

[0130] The database construction unit 103 is used to calculate the increase ratio of the monthly gas sales in the cold winter or the decrease ratio of the monthly gas sales in the warm winter based on the monthly gas sales in the test month in the normal temperature winter and the monthly gas sales corresponding to multiple predicted temperatures in the cold winter or warm winter, and to build a database of the impact of temperature changes on monthly gas sales;

[0131] The calculation unit 104 is used to calculate the monthly gas sales volume for cold or warm winters based on the database of the temperature of the cold winter or warm winter to be predicted, the impact of temperature changes on the monthly gas sales volume, and the monthly gas sales volume for the purpose of predicting the month.

[0132] A third exemplary embodiment of the present invention provides an electronic device.

[0133] Figure 3 An electronic device provided by an embodiment of the present invention is schematically shown.

[0134] Reference Figure 3 As shown, an electronic device 200 provided by an embodiment of the present invention includes: a processor 201, a communication interface 202, a memory 203 and a communication bus 204, wherein the processor 201, the communication interface 202 and the memory 203 communicate with each other through the communication bus 204;

[0135] Memory 203, used for storing programs and / or instructions;

[0136] Processor 201, configured to execute the program and / or instruction;

[0137] Wherein, when executing the program and / or instruction, the processor 201 implements the steps of the method for predicting monthly gas consumption and sales volume of cold and warm winter city gas as described above.

[0138] The fourth exemplary embodiment of the present invention further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned method for predicting monthly gas consumption and sales volume of cold and warm winter city gas are implemented.

[0139] The computer-readable storage medium may be included in the device / apparatus described in the above embodiment; or it may exist independently without being assembled into the device / apparatus. The computer-readable storage medium carries one or more programs, and when the one or more programs are executed, the method according to the embodiment of the present invention is implemented.

[0140] According to an embodiment of the present invention, the computer-readable storage medium may be a non-volatile computer-readable computer-readable storage medium, for example, may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, apparatus, or device.

[0141] Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent substitutions for some of the technical features therein; and these modifications or substitutions do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting monthly city gas consumption and sales volume in cold and warm winters, characterized in that: include: Based on the annual total natural gas demand and the daily unevenness coefficient in winter with normal temperature, the monthly gas sales volume in the test month in winter with normal temperature is obtained, wherein the daily unevenness coefficient is defined as daily gas consumption / average daily gas consumption, and the average daily gas consumption is defined as annual gas sales volume / 365 days; Based on multiple predicted temperatures of cold winter or warm winter and the relationship curve between daily degree unevenness coefficient and temperature, the monthly gas sales volume corresponding to multiple predicted temperatures of cold winter or warm winter is obtained; Based on the monthly gas sales of the test month in a normal winter and the monthly gas sales corresponding to multiple predicted temperatures in a cold winter or a warm winter, the increase ratio of the monthly gas sales in a cold winter or the decrease ratio of the monthly gas sales in a warm winter is calculated, and a database of the impact of temperature changes on monthly gas sales is established; Based on the temperature of the month to be predicted, the database of the impact of temperature changes on monthly gas sales, and the non-final predicted monthly gas sales of the predicted month, the final predicted monthly gas sales for cold and warm winters are calculated.

2. The method for predicting monthly city gas consumption and sales volume in cold and warm winters according to claim 1 is characterized in that: The total annual natural gas demand is predicted by the grey GM (1, 1) model.

3. The method for predicting monthly city gas consumption and sales volume in cold and warm winters according to claim 1 is characterized in that: The monthly gas sales volume of the test month in winter with normal temperature is obtained based on the annual total natural gas demand and the daily uneven coefficient in winter with normal temperature, including: Calculate the product of the total annual natural gas demand and the daily uneven coefficient in winter with normal temperature to obtain the daily gas sales volume in winter under normal conditions; The daily gas sales corresponding to different dates in the test month in winter with normal temperature are added up to obtain the monthly gas sales of the test month in winter with normal temperature.

4. The method for predicting monthly city gas consumption and sales volume in cold and warm winters according to claim 1 is characterized in that: The determination of the plurality of predicted temperatures comprises: Based on the historical average daily temperature in winter, a plurality of predicted temperatures for a cold winter is determined by a decreasing gradient method, or a plurality of predicted temperatures for a warm winter is determined by an increasing gradient method.

5. The method for predicting monthly city gas consumption and sales volume in cold and warm winters according to claim 1 is characterized in that: The acquisition of the relationship curve between the daily unevenness coefficient and the temperature includes: Based on the historical data of urban temperature and historical natural gas sales data, the benchmark temperature is screened out; According to the historical sales data of natural gas and the benchmark temperature, the relationship matrix between the daily degree unevenness coefficient and the benchmark temperature is constructed; Based on the relationship matrix between daily degree unevenness coefficient and reference temperature, the abnormal points of the relationship between daily degree unevenness coefficient and reference temperature are eliminated to obtain the data points of daily degree unevenness coefficient and reference temperature; Based on the daily degree unevenness coefficient and the reference temperature data points, a piecewise linear fitting curve of the relationship between the daily degree unevenness coefficient and the temperature number is performed according to the reference temperature interval.

6. The method for predicting monthly city gas consumption and sales volume in cold and warm winters according to claim 5 is characterized in that: The temperature history data includes the maximum temperature, the minimum temperature and the average temperature of each day, and the natural gas historical sales data includes the natural gas sales of each day.

7. The method for predicting monthly city gas consumption and sales volume in cold and warm winters according to claim 5 is characterized in that: In the step of filtering out the benchmark temperature based on the historical city temperature data and the historical natural gas sales data, the specific operation of filtering is as follows: Compare the correlation coefficients between the highest temperature and natural gas sales, the lowest temperature and natural gas sales, and the average temperature and natural gas sales, and select the temperature with the largest correlation coefficient as the benchmark temperature.

8. The method for predicting monthly city gas consumption and sales volume in cold and warm winters according to claim 5 is characterized in that: The method of obtaining monthly gas sales corresponding to the multiple predicted temperatures of the cold winter or warm winter based on the relationship curve between the multiple predicted temperatures of the cold winter or warm winter and the daily degree unevenness coefficient and the temperature includes: Substituting the specific predicted temperature of cold winter or warm winter into the relationship curve between daily degree unevenness coefficient and temperature, the daily degree unevenness coefficient under the specific predicted temperature of cold winter or warm winter is obtained; Calculate the product of the annual total natural gas demand and the daily unevenness coefficient under the specific predicted temperature of cold winter or warm winter to obtain the daily gas sales volume under the specific predicted temperature of cold winter or warm winter; The daily gas sales volume under the specific predicted temperature of cold winter or warm winter corresponding to different dates in a month is added up to obtain the monthly gas sales volume corresponding to the specific predicted temperature of cold winter or warm winter; The calculation of daily unevenness coefficient, daily gas sales volume and monthly gas sales volume under different predicted temperatures is repeated to obtain monthly gas sales volume corresponding to multiple predicted temperatures of cold winter or warm winter.

9. The method for predicting monthly city gas consumption and sales volume in cold and warm winters according to any one of claims 1 to 8, characterized in that: The monthly gas sales volume of the predicted month is not the final predicted amount, which is the monthly gas sales volume of the month to be tested in winter with normal temperature or the monthly gas sales volume of the month to be tested predicted based on historical monthly gas sales data.

10. A device for predicting monthly city gas consumption and sales volume in cold and warm winters, characterized in that: include: Normal temperature winter prediction unit, cold and warm winter prediction unit, database construction unit and calculation unit; The normal temperature winter prediction unit is used to obtain the monthly gas sales volume of the test month in the normal temperature winter based on the annual total natural gas demand and the daily uneven coefficient of the normal temperature winter, wherein the daily uneven coefficient is defined as the daily gas consumption / average daily gas consumption, and the average daily gas consumption is defined as the annual gas sales volume / 365 days; The cold and warm winter prediction unit is used to obtain monthly gas sales corresponding to multiple predicted temperatures of cold winter or warm winter based on multiple predicted temperatures of cold winter or warm winter and a relationship curve between daily degree unevenness coefficient and temperature; The database construction unit is used to calculate the increase ratio of monthly gas sales in cold winter or the decrease ratio of monthly gas sales in warm winter based on the monthly gas sales in the test month in normal temperature winter and the monthly gas sales corresponding to multiple predicted temperatures in cold winter or warm winter, and to build a database of the impact of temperature changes on monthly gas sales; The calculation unit is used to calculate the final predicted monthly gas sales volume for cold and warm winters based on the temperature of the month to be predicted, the database of the impact of temperature changes on monthly gas sales volume, and the non-final predicted monthly gas sales volume of the predicted month.

11. An electronic device, characterized in that: include: A processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; A memory for storing programs and / or instructions; A processor, configured to execute the program and / or instructions; Wherein, when executing the program and / or instruction, the processor implements the method for predicting monthly gas consumption and sales volume of cold and warm winter city gas as described in any one of claims 1 to 9.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for predicting monthly city gas consumption and sales volume in cold and warm winters according to any one of claims 1-9 is implemented.

Images