Coal water transportation demand prediction method and device based on industrial association

Abstract

The invention provides a coal water transportation demand prediction method and device based on industry association, and relates to the technical field of coal water transportation demand prediction, and the method comprises the steps: screening out a key yield index from the yield indexes related to the port coal throughput in each coal consumption main industry through the calculation of a correlation coefficient; constructing a nonlinear regression model between each key yield index and the port coal throughput; solving a regression coefficient and a corresponding R square of each nonlinear regression model by using historical data; calculating the score of each key yield index based on the coal consumption proportion of all key yield indexes in the current year and the R parties of all nonlinear regression models; and substituting the key yield index of the current year into the corresponding nonlinear regression model, and performing weighted summation on the nonlinear regression model based on the score to obtain the port coal throughput of the next year. The technical problem that an existing port coal throughput prediction method is low in prediction result accuracy is effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of coal water transport demand forecasting, and in particular to a method and device for forecasting coal water transport demand based on industrial association. Background Art

[0002] In order to ensure the smooth operation of various coal-consuming industries, coal must be efficiently transported from the production area to downstream enterprises. However, due to factors such as hydropower output, the amount of coal discharged from ports will fluctuate, which poses a challenge to the arrangement of transportation capacity such as railways and ships, as well as port inventory management. Therefore, accurately predicting the demand for coal transportation at ports is of great significance for the reasonable arrangement of transportation capacity, ensuring the supply of power coal, and guiding the construction of relevant coal terminal berths.

[0003] Traditional port coal throughput forecasting methods mainly rely on historical port coal throughput, and predict future port coal throughput by establishing time series, exponential smoothing and other relationship models, but these methods ignore the impact of factors such as industrial structure adjustment, economic structure transformation, and dynamic changes in transportation intensity on coal demand. Therefore, the existing port coal throughput forecasting methods have the technical problem of low accuracy of forecast results. Summary of the invention

[0004] The purpose of the present invention is to provide a method and device for predicting coal water transport demand based on industrial association, so as to alleviate the technical problem of low accuracy of prediction results existing in the existing port coal throughput prediction method.

[0005] In a first aspect, the present invention provides a method for predicting coal water transport demand based on industrial association, comprising: determining the production index related to the port coal throughput in each main coal-consuming industry; calculating the correlation coefficient between each production index and the port coal throughput, so as to screen out key production indicators from the production index based on the correlation coefficient; constructing a nonlinear regression model between each key production index and the port coal throughput, and solving the regression coefficient and the corresponding R-square of each nonlinear regression model based on historical data; wherein the historical data comprises: multiple groups of key production indicators of the first historical year and the port coal throughput of the second historical year; the first historical year is the previous year of the second historical year; obtaining the coal consumption ratio of each key production indicator in the current year, so as to calculate the score of each key production indicator based on the coal consumption ratio of all key production indicators in the current year and the R-square of all nonlinear regression models; substituting the key production indicators of the current year into the corresponding nonlinear regression model, and performing weighted summation on the corresponding nonlinear regression model based on the score of each key production indicator to obtain the port coal throughput of the next year.

[0006] Optionally, the correlation coefficient includes: a maximum information coefficient; calculating the correlation coefficient between each production indicator and the port coal throughput, so as to screen out key production indicators from the production indicators based on the correlation coefficient, including: calculating the maximum information coefficient between each production indicator and the port coal throughput to obtain multiple maximum information coefficients; when it is determined that the maximum information coefficient of the target production indicator is greater than or equal to a preset threshold, determining the target production indicator as a key production indicator; wherein the target production indicator represents any indicator among all production indicators; when it is determined that the maximum information coefficient of the target production indicator is less than a preset threshold, determining the target production indicator as a non-key production indicator.

[0007] Optionally, solving the regression coefficient and corresponding R-square of each nonlinear regression model based on historical data includes: substituting the target key production indicator of the first historical year and the port coal throughput of the second historical year of each group into the corresponding nonlinear regression model, and obtaining a group of nonlinear regression equations corresponding to the target key production indicator; wherein the target key production indicator represents any indicator among the key production indicators; solving the group of nonlinear regression equations using the least squares method to obtain the regression coefficient of the nonlinear regression model corresponding to the target key production indicator; substituting the regression coefficient into the nonlinear regression model corresponding to the target key production indicator to obtain the target nonlinear regression model; and calculating the R-square of the target nonlinear regression model based on the target nonlinear regression model and historical data.

[0008] Optionally, based on the coal consumption ratio of all key production indicators in the current year and the R-square of all nonlinear regression models, the score of each key production indicator is calculated, including: normalizing the coal consumption ratio of each key production indicator in the current year and the R-square of the corresponding nonlinear regression model, respectively, to obtain a normalized coal consumption ratio set and a normalized R-square set; determining the maximum ratio and the minimum ratio in the normalized coal consumption ratio set, and determining the maximum R-square and the minimum R-square in the normalized R-square set; calculating the optimal solution distance of the target key production indicator based on the normalized coal consumption ratio of the target key production indicator, the normalized R-square, the maximum ratio and the maximum R-square; wherein the target key production indicator represents any indicator among the key production indicators; calculating the worst solution distance of the target key production indicator based on the normalized coal consumption ratio of the target key production indicator, the normalized R-square, the minimum ratio and the minimum R-square; calculating the score of the target key production indicator based on the optimal solution distance and the worst solution distance of the target key production indicator.

[0009] Optionally, a weighted summation is performed on the corresponding nonlinear regression model based on the score of each key yield indicator, including: calculating the sum of the scores of all key yield indicators to obtain a total score; calculating the ratio of the score of the target key yield indicator to the total score to obtain the weight coefficient of the nonlinear regression model corresponding to the target key yield indicator; and using each weight coefficient to perform a weighted summation on the corresponding nonlinear regression model.

[0010] Optionally, the main coal-consuming industries include: electric power industry, metallurgical industry, petrochemical industry and building materials industry.

[0011] Optionally, key production indicators include: thermal power generation, pig iron production, naphtha production, ethylene production, cement production and flat glass production.

[0012] In a second aspect, the present invention provides a coal water transport demand forecasting device based on industrial association, comprising: a determination module, used to determine the production index related to the port coal throughput in each main coal-consuming industry; a calculation and screening module, used to calculate the correlation coefficient between each production index and the port coal throughput, so as to screen out key production indicators from the production index based on the correlation coefficient; a construction and solution module, used to construct a nonlinear regression model between each key production index and the port coal throughput, and solve the regression coefficient and the corresponding R-square of each nonlinear regression model based on historical data; wherein the historical data includes: multiple groups of key production indicators of the first historical year and the port coal throughput of the second historical year; the first historical year is the previous year of the second historical year; an acquisition and calculation module, used to obtain the coal consumption ratio of each key production indicator in the current year, so as to calculate the score of each key production indicator based on the coal consumption ratio of all key production indicators in the current year and the R-square of all nonlinear regression models; a prediction module, used to substitute the key production indicators of the current year into the corresponding nonlinear regression model, and perform weighted summation on the corresponding nonlinear regression model based on the score of each key production indicator to obtain the port coal throughput of the next year.

[0013] In a third aspect, the present invention provides an electronic device including a memory and a processor, wherein the memory stores a computer program executable on the processor, and when the processor executes the computer program, the method for predicting coal water transport demand based on industrial association according to any one of the aforementioned embodiments is implemented.

[0014] In a fourth aspect, the present invention provides a computer-readable storage medium, which stores computer instructions. When the computer instructions are executed by a processor, the method for predicting coal water transportation demand based on industrial association in any of the aforementioned implementation modes is implemented.

[0015] The present invention provides a method for predicting coal water transport demand based on industrial association. The method first selects key production indicators from the production indicators related to the port coal throughput in each main coal consumption industry by calculating the correlation coefficient, then constructs a nonlinear regression model between each key production indicator and the port coal throughput, and uses historical data to solve the regression coefficient and the corresponding R square of each nonlinear regression model, and then objectively and reasonably calculates the score of each key production indicator based on the coal consumption ratio of all key production indicators in the current year and the R square of all nonlinear regression models, and finally substitutes the key production indicators of the current year into the corresponding nonlinear regression model, and performs weighted summation on the corresponding nonlinear regression model based on the score of each key production indicator to obtain the port coal throughput of the next year. The method analyzes the correlation between the production indicators of multiple main coal consumption industries and the coal water transport volume according to the development and changes of the influencing factors behind the coal water transport volume, and constructs a prediction model for coal throughput based on the causal relationship between multiple key production indicators and the port coal throughput. Therefore, the prediction model constructed based on the above method has the advantages of strong objectivity and high accuracy, and can effectively alleviate the technical problem of low accuracy of the prediction results in the existing port coal throughput prediction method. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 A flow chart of a method for predicting coal water transport demand based on industrial association provided by an embodiment of the present invention; Figure 2 It is a fitting result diagram of the nonlinear regression model corresponding to the thermal power generation determined based on historical data; Figure 3 It is a comparison curve chart of the actual value and predicted value of the port's coal throughput; Figure 4 A functional module diagram of a coal water transport demand forecasting device based on industry association provided by an embodiment of the present invention; Figure 5 A schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions 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. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0019] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. 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.

[0020] Some embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0021] The coal throughput of ports is greatly affected by the market. If we only rely on historical throughput forecasts, it will be difficult to be convincing. If we only analyze the relationship between a certain industry and coal demand, there will inevitably be deviations. Coal application services are in many industries, and it is difficult for a single industry to accurately reflect the overall characteristics of the market. If we consider forecasts from multiple industries, it will be difficult to objectively determine the weight coefficients of each industry. In view of this, the embodiment of the present invention provides a method for predicting coal water transport demand based on industrial associations to alleviate the technical problems involved above.

[0022] Embodiment 1 Figure 1 A flow chart of a method for predicting coal water transport demand based on industrial association provided by an embodiment of the present invention, such as Figure 1 As shown, the method specifically comprises the following steps: Step S102, determining the production index related to the port coal throughput in each major coal consuming industry.

[0023] Specifically, the embodiment of the present invention is a method for predicting the port coal throughput based on the causal relationship between the coal consumption of multiple main coal-consuming industries and the port coal throughput. Therefore, it is first necessary to clarify which industries belong to the main coal-consuming industries. Based on the analysis of my country's current coal consumption, the main coal-consuming industries include: power industry, metallurgical industry, petrochemical industry and building materials industry; there may be differences in the main coal-consuming industries in different geographical regions, which can be determined based on the actual coal consumption of industries in each region.

[0024] Next, collect coal consumption data for each major coal-consuming industry within a certain time frame. These data can be obtained from the Statistics Bureau, industry association reports, corporate annual reports, and public market research reports. At the same time, it is also necessary to collect coal throughput data for relevant ports. These data can come from official statistical reports of ports, customs data, and statistics released by other third-party agencies. Then, analyze the collected data to determine the various production indicators associated with port coal throughput.

[0025] Step S104, calculating the correlation coefficient between each production index and the port coal throughput, so as to screen out key production indicators from the production indicators based on the correlation coefficient.

[0026] It is known that there are a large number of production indicators related to port coal throughput. However, considering the balance between calculation amount and prediction accuracy, the embodiment of the present invention chooses to first screen out key production indicators from multiple production indicators, and then predict the port coal throughput based on the key production indicators.

[0027] Specifically, by calculating the correlation coefficient between each production index and the port coal throughput, it is found out which production index has the highest correlation with the port coal throughput, so that the production index with a correlation coefficient exceeding a certain threshold is used as the key production index. The embodiment of the present invention does not specifically limit the correlation measurement method between the production index and the port coal throughput, for example, Pearson correlation coefficient, Spearman rank correlation coefficient, Maximum Information Coefficient (MIC).

[0028] Based on the correlation analysis between various production indicators and port coal throughput in my country's main coal consumption industries, it can be determined that the key production indicators include: thermal power generation, pig iron production, naphtha production, ethylene production, cement production and flat glass production. Among them, thermal power generation is the production indicator of the power industry, pig iron production is the production indicator of the metallurgical industry, naphtha production and ethylene production are both production indicators of the petrochemical industry, and cement production and flat glass production are both production indicators of the building materials industry.

[0029] Step S106, constructing a nonlinear regression model between each key production indicator and the port coal throughput, and solving the regression coefficient and corresponding R-square of each nonlinear regression model based on historical data.

[0030] After determining multiple key production indicators, the embodiment of the present invention creates a corresponding nonlinear regression model for each key production indicator. In the embodiment of the present invention, the expression of the nonlinear regression model is as follows: ,in, represents the regression coefficient of the nonlinear regression model, Represents the key production indicator, Represents the port coal throughput. Taking the key output indicator as thermal power generation as an example, the corresponding nonlinear regression model is established with thermal power generation as the input variable and port coal throughput as the output variable.

[0031] If the number of key production indicators is N, then the number of nonlinear regression models is N. To use the nonlinear regression model to predict the port coal throughput, it is obviously necessary to first solve the regression coefficient of each nonlinear regression model. The embodiment of the present invention substitutes multiple sets of historical data of key production indicators and port coal throughput into the corresponding nonlinear regression model, that is, substitutes multiple sets of data into the model, and solves the regression coefficient of each nonlinear regression model through statistical methods. In addition, the embodiment of the present invention also calculates the R square of the model based on the determined nonlinear regression model to measure the closeness between the model predicted data and the actual data. Among them, the historical data includes: multiple sets of key production indicators in the first historical year and the port coal throughput in the second historical year; the first historical year is the previous year of the second historical year.

[0032] Step S108, obtaining the coal consumption ratio of each key production indicator in the current year, and calculating the score of each key production indicator based on the coal consumption ratio of all key production indicators in the current year and the R-square of all nonlinear regression models.

[0033] In view that multiple key production indicators of multiple industries can predict the coal throughput of ports, it is necessary to determine the respective weight coefficients for the prediction results of multiple nonlinear regression models that have been constructed. Therefore, the embodiment of the present invention chooses to calculate the score of each key production indicator according to the coal consumption ratio of all key production indicators in the current year and the R-square of all nonlinear regression models, and then determines the weight coefficient of the prediction result of the corresponding nonlinear regression model according to its score. In the embodiment of the present invention, the higher the score of the key production indicator, the greater the weight coefficient of the prediction result of the corresponding nonlinear regression model.

[0034] Step S110, substituting the key production indicators of the current year into the corresponding nonlinear regression model, and performing weighted summation on the corresponding nonlinear regression model based on the score of each key production indicator to obtain the port coal throughput of the next year.

[0035] In the embodiment of the present invention, the prediction model of the port coal throughput is the result of weighted summation of multiple key production indicators to the corresponding nonlinear regression model, that is, ,in, represents the forecast result of the port coal throughput, According to key production indicators Forecasted port coal throughput, , According to The rating is determined The weight coefficient of n represents the total number of key production indicators.

[0036] To determine the port coal throughput for the next year, the key production indicators of the current year are substituted into the respective nonlinear regression models, and the multiple port coal throughputs output by multiple models are weighted according to the corresponding weight coefficients and then summed up to obtain the predicted results of the port coal throughput.

[0037] The present invention provides a method for predicting coal water transport demand based on industrial association. The method first selects key production indicators from the production indicators related to the port coal throughput in each main coal consumption industry by calculating the correlation coefficient, then constructs a nonlinear regression model between each key production indicator and the port coal throughput, and uses historical data to solve the regression coefficient and the corresponding R square of each nonlinear regression model, and then objectively and reasonably calculates the score of each key production indicator based on the coal consumption ratio of all key production indicators in the current year and the R square of all nonlinear regression models, and finally substitutes the key production indicators of the current year into the corresponding nonlinear regression model, and performs weighted summation on the corresponding nonlinear regression model based on the score of each key production indicator to obtain the port coal throughput of the next year. The method analyzes the correlation between the production indicators of multiple main coal consumption industries and the coal water transport volume according to the development and changes of the influencing factors behind the coal water transport volume, and constructs a prediction model for coal throughput based on the causal relationship between multiple key production indicators and the port coal throughput. Therefore, the prediction model constructed based on the above method has the advantages of strong objectivity and high accuracy, and can effectively alleviate the technical problem of low accuracy of the prediction results in the existing port coal throughput prediction method.

[0038] In an optional implementation, the correlation coefficient includes: a maximum information coefficient; the above step S104, calculating the correlation coefficient between each production index and the port coal throughput, so as to screen out the key production index from the production index based on the correlation coefficient, specifically includes the following steps: Step S1041, calculating the maximum information coefficient between each production index and the port coal throughput, and obtaining multiple maximum information coefficients.

[0039] Step S1042, when it is determined that the maximum information coefficient of the target production indicator is greater than or equal to a preset threshold, the target production indicator is determined to be a key production indicator; wherein the target production indicator represents any indicator among all production indicators.

[0040] Step S1043, when it is determined that the maximum information coefficient of the target production indicator is less than a preset threshold, the target production indicator is determined to be a non-critical production indicator.

[0041] Specifically, the maximum information coefficient is a method for measuring the correlation between two variables, which has strong universality, fairness and symmetry. Therefore, the embodiment of the present invention selects the calculation results of the maximum information coefficient between each production index and the port coal throughput, selects the key production index, and takes the production index with the maximum information coefficient greater than or equal to the preset threshold as the key production index. The maximum information coefficient is a prior art method in the field of statistics, so the embodiment of the present invention will not further describe its calculation process.

[0042] In an optional implementation, in the above step S106, solving the regression coefficient and the corresponding R-square of each nonlinear regression model based on historical data specifically includes the following steps: Step S1061, substitute each group of target key production indicators of the first historical year and the port coal throughput of the second historical year into the corresponding nonlinear regression model respectively to obtain a group of nonlinear regression equations corresponding to the target key production indicators; wherein the target key production indicator represents any indicator among the key production indicators.

[0043] Step S1062, using the least squares method to solve the nonlinear regression equation group, to obtain the regression coefficient of the nonlinear regression model corresponding to the target key production index.

[0044] Step S1063, substituting the regression coefficient into the nonlinear regression model corresponding to the target key production index to obtain the target nonlinear regression model.

[0045] Step S1064, calculating the R-square of the target nonlinear regression model based on the target nonlinear regression model and historical data.

[0046] For ease of understanding, the above steps are explained below using the nonlinear regression model corresponding to thermal power generation as an example. Assuming that the data from 2009 to 2017 are historical data, the thermal power generation in 2009 and the port coal throughput in 2010 are a set of input and output data for the model, the thermal power generation in 2010 and the port coal throughput in 2011 are a set of input and output data for the model, and so on, multiple sets of input and output data for the nonlinear regression model are obtained.

[0047] After each set of input and output data is substituted into the nonlinear regression model corresponding to thermal power generation, an unknown number is obtained as 3 regression coefficients If the above multiple input and output groups are substituted into the nonlinear regression model respectively, the nonlinear regression equation group corresponding to the thermal power generation can be obtained.

[0048] The embodiment of the present invention adopts the least square method to solve the above nonlinear regression equations to determine the regression coefficient of the nonlinear regression model corresponding to the thermal power generation. After the regression coefficient is determined, it is substituted into the nonlinear regression model corresponding to the thermal power generation to obtain the target nonlinear regression model corresponding to the thermal power generation.

[0049] Next, if the thermal power generation in 2009 in the historical data is input into the target nonlinear regression model, the port coal throughput predicted by the model for 2010 (that is, the predicted value) can be obtained. Based on multiple sets of predicted values ​​and the corresponding actual values ​​in the historical data, the R-square of the nonlinear regression model can be calculated.

[0050] In order to accurately predict the port coal throughput in future years, the time span between the years corresponding to the historical data used to solve each nonlinear regression model should be as small as possible. For example, to predict the port coal throughput in 2023, it is preferred to use historical data from 2022 to several previous years (such as 2010) for the solution.

[0051] Table 1 below is an example of historical data. If the data from 2009 to 2017 is used as an example to solve the nonlinear regression model of each key production indicator, then the nonlinear regression model corresponding to thermal power generation is: ,in, represents the amount of thermal power generation, It represents the port coal throughput predicted based on thermal power generation, and the R-square of the model is: . Figure 2 This is the fitting result diagram of the nonlinear regression model corresponding to the thermal power generation determined based on historical data. Figure 2 It can be seen that the fitting effect of the model is good.

[0052] Table 1 Historical data

[0053] Similarly, the nonlinear regression model corresponding to pig iron production is: ,in, represents the output of pig iron, represents the port coal throughput predicted based on pig iron production, and the R-square of the model is: .

[0054] The nonlinear regression model corresponding to naphtha production is: ,in, represents the naphtha production, represents the port coal throughput predicted based on naphtha production, and the R-square of the model is: .

[0055] The nonlinear regression model corresponding to ethylene production is: ,in, represents the ethylene production, represents the port coal throughput predicted based on ethylene production, and the R-square of the model is: .

[0056] The nonlinear regression model corresponding to cement production is: ,in, represents cement production, represents the port coal throughput predicted based on cement production, and the R-square of the model is: .

[0057] The nonlinear regression model corresponding to the output of flat glass is: ,in, represents the output of flat glass, represents the port coal throughput predicted based on the flat glass production, and the R-square of the model is: .

[0058] In an optional implementation, in the above step S108, based on the coal consumption ratio of all key production indicators in the current year and the R-square of all nonlinear regression models, the score of each key production indicator is calculated, which specifically includes the following steps: Step S1081, respectively normalize the coal consumption ratio of each key output indicator in the current year and the R-square of the corresponding nonlinear regression model to obtain a normalized coal consumption ratio set and a normalized R-square set.

[0059] Table 2 below shows the coal consumption ratio of each key output indicator in the current year and the R-squared data of the corresponding nonlinear regression model. For the coal consumption ratio and the R-squared, the embodiment of the present invention uses extremely large indicators for normalization processing, and the expression of the normalization processing is: ,in, represents the index before normalization, Indicates the maximum value of the indicator. Indicates the minimum value of the indicator. Represents the normalized index.

[0060] Table 2 Coal consumption ratio of key production indicators and R-squared of corresponding models

[0061] Taking the coal consumption ratio as an example, first determine the maximum and minimum values ​​of the coal consumption ratio. Then, to normalize the coal consumption ratio of a key output indicator in the current year, substitute it into z in the above formula, and substitute the maximum and minimum values ​​of the coal consumption ratio into z in the above formula respectively. and , and then the normalized coal consumption ratio of the key output indicator in the current year can be obtained. The normalization method of R-square of each nonlinear regression model is similar, so it will not be repeated here.

[0062] Step S1082, determining the maximum ratio and the minimum ratio in the normalized coal consumption ratio set, and determining the maximum R-square and the minimum R-square in the normalized R-square set.

[0063] According to the normalized coal consumption ratio set and the normalized R-square set, the index vector can be constructed ,in, It represents the normalized coal consumption ratio of the nth key production indicator in the current year, It represents the normalized R-square of the nonlinear regression model corresponding to the nth key production indicator. Based on this, the optimal solution vector based on the maximum proportion and the maximum R-square is expressed as: ; The worst solution vector based on the minimum ratio and the minimum R square is expressed as: .

[0064] Step S1083, based on the normalized coal consumption ratio, normalized R-square, maximum ratio and maximum R-square of the target key production indicator, calculate the optimal solution distance of the target key production indicator; wherein the target key production indicator represents any indicator among the key production indicators.

[0065] In the embodiment of the present invention, the formula for the optimal solution distance of the target key production indicator i is: ;in, Indicates the maximum ratio, represents the normalized coal consumption ratio of the target key output indicator i, represents the maximum R-square, It represents the normalized R-square of the nonlinear regression model corresponding to the target key production indicator i. Represents the optimal solution distance of the target key production indicator i.

[0066] Step S1084, based on the normalized coal consumption ratio of the target key production index, the normalized R-square, the minimum ratio and the minimum R-square, calculate the worst solution distance of the target key production index.

[0067] In the embodiment of the present invention, the formula for the worst solution distance of the target key yield indicator i is: ;in, Indicates the minimum ratio, represents the minimum R-squared, Represents the worst solution distance of the target key output indicator i.

[0068] Step S1085, based on the optimal solution distance and the worst solution distance of the target key yield indicator, calculate the score of the target key yield indicator.

[0069] Specifically, the scoring formula for the target key output indicator i is: .

[0070] In an optional implementation, in the above step S110, weighted summing of the corresponding nonlinear regression models is performed based on the score of each key yield indicator, specifically including the following steps: Step S1101, calculate the sum of the scores of all key production indicators to obtain a total score.

[0071] Step S1102, calculating the ratio of the score of the target key yield indicator to the total score, and obtaining the weight coefficient of the nonlinear regression model corresponding to the target key yield indicator.

[0072] Based on the above Table 2, the above method can be used to calculate the score of each key yield indicator, and then calculate the weight coefficient of the nonlinear regression model corresponding to each key yield indicator. For details, please refer to the following Table 3.

[0073] Table 3 Weight coefficients of key production indicators

[0074] Step S1103: perform weighted summation on the corresponding nonlinear regression model using each weight coefficient.

[0075] When predicting the port coal throughput, after solving the coefficients of the nonlinear regression model through historical data, the key production indicators of the current year are substituted into the respective models to solve the port coal throughput predicted by each model, and then each is multiplied by the corresponding weight coefficient, and the port coal throughput of the next year is obtained by weighted summation, that is, .

[0076] In summary, the embodiment of the present invention determines the key production indicators related to the port coal throughput through the maximum information coefficient, and uses the nonlinear regression model to establish the correlation between the port coal throughput and each key production indicator, and then solves each nonlinear regression model through historical data. In addition, the embodiment of the present invention uses the coal consumption ratio of all key production indicators in the current year and the R square of all nonlinear regression models to calculate the distance between each key production indicator and the optimal solution and the worst solution, so as to objectively and reasonably determine the weight coefficient of each key production indicator, and finally weight multiple nonlinear regression models based on the above weight coefficient to construct a prediction model for port coal throughput. This prediction model can carry out predictions for different scenarios, predict my country's port coal throughput under the conditions of different product outputs in different industries, and provide scientific data support for macroeconomic and industry operation monitoring and analysis.

[0077] In order to verify the effectiveness of this method, the historical data from 2009 to 2017 were used as sample data for fitting to form a prediction model for port coal throughput. The predicted water transport coal demand data for previous years was compared with the actual data. The calculation results showed that the average error between the model fitting value and the actual value from 2009 to 2017 was 0.2%. The prediction effect is good. Figure 3 This is a comparison curve between the actual value and the predicted value of the port's coal throughput.

[0078] Embodiment 2 An embodiment of the present invention also provides a device for predicting coal water transportation demand based on industrial association, which is mainly used to execute the method for predicting coal water transportation demand based on industrial association provided in the above-mentioned embodiment one. The following is a specific introduction to the device for predicting coal water transportation demand based on industrial association provided in an embodiment of the present invention.

[0079] Figure 4 A functional module diagram of a coal water transport demand forecasting device based on industry association provided by an embodiment of the present invention, such as Figure 4 As shown, the device mainly includes: a determination module 10, a calculation and screening module 20, a construction and solution module 30, an acquisition and calculation module 40, and a prediction module 50, wherein: The determination module 10 is used to determine the production index related to the port coal throughput in each major coal consuming industry.

[0080] The calculation and screening module 20 is used to calculate the correlation coefficient between each production index and the port coal throughput, so as to screen out key production indicators from the production indicators based on the correlation coefficient.

[0081] The construction and solution module 30 is used to construct a nonlinear regression model between each key production indicator and the port coal throughput, and solve the regression coefficient and the corresponding R-square of each nonlinear regression model based on historical data; wherein the historical data includes: multiple groups of key production indicators in the first historical year and the port coal throughput in the second historical year; the first historical year is the previous year of the second historical year.

[0082] The acquisition and calculation module 40 is used to obtain the coal consumption ratio of each key production indicator in the current year, so as to calculate the score of each key production indicator based on the coal consumption ratio of all key production indicators in the current year and the R-square of all nonlinear regression models.

[0083] The prediction module 50 is used to substitute the key production indicators of the current year into the corresponding nonlinear regression model, and perform weighted summation on the corresponding nonlinear regression model based on the score of each key production indicator to obtain the port coal throughput of the next year.

[0084] The embodiment of the present invention provides a coal water transport demand prediction device based on industrial association. The device first selects key production indicators from the production indicators related to the port coal throughput in each main coal consumption industry by calculating the correlation coefficient, then constructs a nonlinear regression model between each key production indicator and the port coal throughput, and uses historical data to solve the regression coefficient and the corresponding R square of each nonlinear regression model, and then objectively and reasonably calculates the score of each key production indicator based on the coal consumption proportion of all key production indicators in the current year and the R square of all nonlinear regression models, and finally substitutes the key production indicators of the current year into the corresponding nonlinear regression model, and performs weighted summation on the corresponding nonlinear regression model based on the score of each key production indicator to obtain the port coal throughput of the next year. The method analyzes the correlation between the production indicators of multiple main coal consumption industries and the coal water transport volume according to the development and changes of the influencing factors behind the coal water transport volume, and constructs a prediction model for coal throughput based on the causal relationship between multiple key production indicators and the port coal throughput. Therefore, the prediction model constructed based on the above method has the advantages of strong objectivity and high accuracy, and can effectively alleviate the technical problem of low accuracy of the prediction results in the existing port coal throughput prediction method.

[0085] Optionally, the correlation coefficient includes: a maximum information coefficient; the calculation and screening module 20 is specifically used for: The correlation coefficient between each production index and the port coal throughput is calculated to screen out key production indicators from the production index based on the correlation coefficient, including: The maximum information coefficient between each production index and the port coal throughput is calculated, and multiple maximum information coefficients are obtained.

[0086] When it is determined that the maximum information coefficient of the target production indicator is greater than or equal to a preset threshold, the target production indicator is determined to be a key production indicator; wherein the target production indicator represents any indicator among all production indicators.

[0087] When it is determined that the maximum information coefficient of the target yield indicator is less than a preset threshold, the target yield indicator is determined to be a non-critical yield indicator.

[0088] The construction and solution module 30 is specifically used for: Substitute the target key production index of the first historical year and the port coal throughput of the second historical year of each group into the corresponding nonlinear regression model respectively to obtain the nonlinear regression equation group corresponding to the target key production index; wherein the target key production index represents any indicator among the key production indicators.

[0089] The least squares method is used to solve the nonlinear regression equations to obtain the regression coefficients of the nonlinear regression model corresponding to the target key production indicators.

[0090] Substitute the regression coefficient into the nonlinear regression model corresponding to the target key yield indicator to obtain the target nonlinear regression model.

[0091] Based on the target nonlinear regression model and historical data, calculate the R-squared of the target nonlinear regression model.

[0092] The acquisition and calculation module 40 is specifically used for: The coal consumption ratio of each key output indicator in the current year and the R-square of the corresponding nonlinear regression model are normalized respectively to obtain the normalized coal consumption ratio set and the normalized R-square set.

[0093] Determine the maximum and minimum proportions in the normalized coal consumption proportion set, and determine the maximum and minimum R-squared in the normalized R-squared set.

[0094] Based on the normalized coal consumption ratio, normalized R-square, maximum ratio and maximum R-square of the target key production index, the optimal solution distance of the target key production index is calculated; wherein the target key production index represents any indicator among the key production indicators.

[0095] Based on the normalized coal consumption ratio of the target key production index, the normalized R-square, the minimum ratio and the minimum R-square, the worst solution distance of the target key production index is calculated.

[0096] Based on the optimal solution distance and the worst solution distance of the target key yield indicator, the score of the target key yield indicator is calculated.

[0097] Optionally, the prediction module 50 is specifically used for: The scores of all key yield indicators are summed to obtain a total score.

[0098] The ratio of the score of the target key yield indicator to the total score is calculated to obtain the weight coefficient of the nonlinear regression model corresponding to the target key yield indicator.

[0099] Each weight coefficient is used to perform weighted summation on the corresponding nonlinear regression model.

[0100] Optionally, the main coal consuming industries include: electric power industry, metallurgical industry, petrochemical industry and building materials industry.

[0101] Optionally, key production indicators include: thermal power generation, pig iron production, naphtha production, ethylene production, cement production and flat glass production.

[0102] Embodiment 3 See also Figure 5 An embodiment of the present invention provides an electronic device, which includes: a processor 60, a memory 61, a bus 62 and a communication interface 63, wherein the processor 60, the communication interface 63 and the memory 61 are connected via the bus 62; the processor 60 is used to execute an executable module stored in the memory 61, such as a computer program.

[0103] The memory 61 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 63 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0104] The bus 62 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0105] Among them, the memory 61 is used to store programs, and the processor 60 executes the program after receiving the execution instruction. The method executed by the device defined by the process disclosed in any embodiment of the above-mentioned embodiment of the present invention can be applied to the processor 60 or implemented by the processor 60.

[0106] The processor 60 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 60. The above processor 60 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in the embodiments of the present invention can be directly embodied as a hardware decoding processor to be executed, or the hardware and software modules in the decoding processor can be executed. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 61, and the processor 60 reads the information in the memory 61 and completes the steps of the above method in combination with its hardware.

[0107] A computer program product of a method and apparatus for predicting coal water transport demand based on industrial association provided in an embodiment of the present invention includes a computer-readable storage medium storing non-volatile program code executable by a processor, and the instructions included in the program code can be used to execute the method described in the previous method embodiment. The specific implementation can be found in the method embodiment, which will not be repeated here.

[0108] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0109] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that can be executed by a processor. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., various media that can store program codes.

[0110] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0111] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inside", "outside", etc. indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, or the positions or positional relationships in which the product of the invention is usually placed when in use. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific position, be constructed and operated in a specific position, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", "third", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0112] In addition, the terms "horizontal", "vertical", "overhanging" and the like do not mean that the components are required to be absolutely horizontal or overhanging, but can be slightly tilted. For example, "horizontal" only means that its direction is more horizontal than "vertical", and does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0113] In the description of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "set", "install", "connect", and "connect" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0114] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. 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 replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting coal water transport demand based on industrial association, characterized in that: include: Determine the production indicators related to port coal throughput in each major coal-consuming industry; Calculating the correlation coefficient between each of the production indicators and the coal throughput of the port, so as to screen out key production indicators from the production indicators based on the correlation coefficient; Constructing a nonlinear regression model between each of the key production indicators and the port coal throughput, and solving the regression coefficient and the corresponding R-square of each of the nonlinear regression models based on historical data; wherein the historical data includes: multiple groups of key production indicators in the first historical year and the port coal throughput in the second historical year; the first historical year is the previous year of the second historical year; Obtaining the coal consumption ratio of each of the key production indicators in the current year, and calculating the score of each of the key production indicators based on the coal consumption ratios of all key production indicators in the current year and the R-square of all nonlinear regression models; The key production indicators of the current year are substituted into the corresponding nonlinear regression model, and the corresponding nonlinear regression model is weighted summed based on the score of each key production indicator to obtain the port coal throughput of the next year.

2. The method for predicting coal water transport demand based on industrial association according to claim 1 is characterized in that: The correlation coefficients include: maximum information coefficient; Calculating the correlation coefficient between each of the production indicators and the coal throughput of the port, so as to screen out key production indicators from the production indicators based on the correlation coefficient, including: Calculating the maximum information coefficient between each of the production indicators and the coal throughput of the port to obtain multiple maximum information coefficients; When it is determined that the maximum information coefficient of the target production index is greater than or equal to a preset threshold, the target production index is determined to be a key production index; wherein the target production index represents any one of all production indexes; When it is determined that the maximum information coefficient of the target yield indicator is less than the preset threshold, the target yield indicator is determined to be a non-critical yield indicator.

3. The method for predicting coal water transport demand based on industrial association according to claim 1 is characterized in that: Solving the regression coefficient and the corresponding R-square of each nonlinear regression model based on historical data includes: Substituting the target key production index of the first historical year and the port coal throughput of the second historical year of each group into the corresponding nonlinear regression model, respectively, to obtain a nonlinear regression equation group corresponding to the target key production index; wherein the target key production index represents any one of the key production indexes; Solving the nonlinear regression equations by the least square method to obtain the regression coefficients of the nonlinear regression model corresponding to the target key production index; Substituting the regression coefficient into the nonlinear regression model corresponding to the target key yield indicator to obtain a target nonlinear regression model; Based on the target nonlinear regression model and the historical data, the R-square of the target nonlinear regression model is calculated.

4. The method for predicting coal water transport demand based on industrial association according to claim 1 is characterized in that: Based on the proportion of coal consumption in the current year and the R-square of all nonlinear regression models, the score of each key production indicator is calculated, including: Normalize the coal consumption ratio of each key output indicator in the current year and the R-square of the corresponding nonlinear regression model to obtain a normalized coal consumption ratio set and a normalized R-square set; Determine the maximum ratio and the minimum ratio in the normalized coal consumption ratio set, and determine the maximum R-square and the minimum R-square in the normalized R-square set; Based on the normalized coal consumption ratio of the target key production index, the normalized R-square, the maximum ratio and the maximum R-square, the optimal solution distance of the target key production index is calculated; wherein the target key production index represents any one of the key production indicators; Calculate the worst solution distance of the target key production index based on the normalized coal consumption ratio of the target key production index, the normalized R-square, the minimum ratio and the minimum R-square; Based on the optimal solution distance and the worst solution distance of the target key production indicator, the score of the target key production indicator is calculated.

5. The method for predicting coal water transport demand based on industrial association according to claim 4 is characterized in that: Based on the score of each of the key yield indicators, a weighted sum is performed on the corresponding nonlinear regression model, including: Calculate the sum of the scores of all key yield indicators to get the total score; Calculating the ratio of the score of the target key yield indicator to the total score to obtain the weight coefficient of the nonlinear regression model corresponding to the target key yield indicator; Each weight coefficient is used to perform weighted summation on the corresponding nonlinear regression model.

6. The method for predicting coal water transport demand based on industrial association according to claim 1 is characterized in that: The main coal-consuming industries include: electric power industry, metallurgical industry, petrochemical industry and building materials industry.

7. The method for predicting coal water transport demand based on industrial association according to claim 1 is characterized in that: The key production indicators include: thermal power generation, pig iron production, naphtha production, ethylene production, cement production and flat glass production.

8. A coal water transport demand forecasting device based on industrial association, characterized in that: include: A determination module for determining the production indicators related to the port coal throughput in each major coal-consuming industry; A calculation and screening module, used for calculating the correlation coefficient between each of the production indicators and the coal throughput of the port, so as to screen out key production indicators from the production indicators based on the correlation coefficient; A construction and solution module, used for constructing a nonlinear regression model between each of the key production indicators and the port coal throughput, and solving the regression coefficient and the corresponding R-square of each of the nonlinear regression models based on historical data; wherein the historical data includes: multiple groups of key production indicators in a first historical year and port coal throughput in a second historical year; the first historical year is the previous year of the second historical year; An acquisition and calculation module, used to acquire the coal consumption ratio of each of the key production indicators in the current year, so as to calculate the score of each of the key production indicators based on the coal consumption ratio of all the key production indicators in the current year and the R-square of all the nonlinear regression models; The prediction module is used to substitute the key production indicators of the current year into the corresponding nonlinear regression model, and perform weighted summation on the corresponding nonlinear regression model based on the score of each key production indicator to obtain the port coal throughput of the next year.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, wherein: When the processor executes the computer program, the method for predicting coal water transportation demand based on industrial association according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the method for predicting coal water transportation demand based on industrial association described in any one of claims 1 to 7 is implemented.

Images