Coal price prediction method and system, electronic equipment and storage medium
By combining ARAM and LSTM models to perform linear and nonlinear partial prediction of coal price time series and optimize model parameters, the problem of coal price fluctuations affecting the economic benefits of power generation enterprises is solved, and accurate coal price prediction and coal purchasing strategy optimization are achieved.
Patent Information
- Application Number
- CN202510509673.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-22
- Publication Date
- 2025-09-05
AI Technical Summary
In the prior art, coal price fluctuations affect the economic benefits of power generation enterprises, and there is a lack of effective price prediction methods, resulting in poor coal purchasing strategies.
The method of combining ARAM model and LSTM model is used to predict the linear and nonlinear parts of the coal price time series respectively, and the model parameters are adjusted through error optimization to achieve accurate prediction of coal prices.
It improves the accuracy and stability of coal price forecasts, helps power generation companies to formulate more effective coal purchasing strategies and improve economic benefits.
Smart Images
Figure CN120598587A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of coal price forecasting, and more specifically, to a coal price forecasting method, system, electronic equipment and storage medium. Background Art
[0002] Coal-fired power generation remains the primary power generation method, and will likely remain so for a long time to come. In the coal market, coal prices are subject to certain fluctuations. As a major component of power generation costs for power generation companies, coal prices significantly impact their economic performance. By integrating modern information technology and understanding the dynamics of coal prices, companies can better formulate coal purchasing strategies, maximize profits, and gain a foothold in the market. Summary of the Invention
[0003] The present invention aims to solve the technical problems existing in the prior art and provides a coal price prediction method, system, electronic equipment and storage medium.
[0004] According to a first aspect of the present invention, a coal price forecasting method is provided, comprising: Obtaining historical prices of target coal to form an actual price time series of the target coal, screening influencing factors on the price time series, and using the price time series and the screened influencing factors to form an initial time series; Predicting the linear portion of the initial time series based on the ARAM model to obtain a predicted value of the linear portion; Stripping the linear portion from the initial time series to obtain a nonlinear portion of the initial time series; Predicting the nonlinear part based on the LSTM model to obtain a predicted value of the nonlinear part; Adding the linear part prediction value and the nonlinear prediction value to obtain a price prediction time series value; Comparing the price prediction time series value with the actual price time series to obtain an error series; Optimizing and adjusting the ARAM model and the LSTM model based on the error sequence to obtain the optimized and adjusted ARAM model and the LSTM model; Based on the optimized and adjusted ARAM model and the LSTM model, the sequence of influencing factors to be predicted is predicted to obtain the coal price.
[0005] According to a second aspect of the present invention, there is provided a coal price forecasting system, comprising: An acquisition module is used to acquire historical prices of target coal to form an actual price time series of the target coal, screen influencing factors of the price time series, and the price time series and the screened influencing factors form an initial time series; a first prediction module, configured to predict the linear portion of the initial time series based on the ARAM model to obtain a linear portion prediction value; and to separate the linear portion from the initial time series to obtain a nonlinear portion of the initial time series, and to predict the nonlinear portion based on the LSTM model to obtain a nonlinear portion prediction value; and to add the linear portion prediction value and the nonlinear prediction value to obtain a price prediction time series value; A calculation module, used to calculate the error sequence between the price prediction time series value and the actual price time series; An optimization module, configured to optimize and adjust the ARAM model and the LSTM model based on the error sequence to obtain the optimized ARAM model and the LSTM model; The second prediction module is used to predict the sequence of influencing factors to be predicted based on the optimized and adjusted ARAM model and the LSTM model to obtain the coal price.
[0006] According to a third aspect of the present invention, there is provided an electronic device comprising a memory and a processor, wherein the processor is configured to implement the steps of a coal price prediction method when executing a computer management program stored in the memory.
[0007] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, on which a computer management program is stored, and when the computer management program is executed by a processor, the steps of the coal price prediction method are implemented.
[0008] The present invention provides a coal price forecasting method, system, electronic device, and storage medium. The method obtains the historical price of a target coal to form a time series of the target coal's actual price. The price time series is screened for influencing factors to form an initial time series. The linear portion of the initial time series is predicted based on an ARAM model, and the nonlinear portion is predicted based on an LSTM model to obtain a predicted value for the nonlinear portion. The predicted price time series value is compared with the actual price time series to obtain an error sequence. The ARAM model and the LSTM model are optimized and adjusted based on the error sequence. The optimized and adjusted ARAM model and LSTM model are used to predict the sequence of influencing factors to obtain the coal price. The present invention predicts the linear and nonlinear portions of the time series based on the ARAM model and the LSTM model, respectively, to achieve trend forecasting of coal prices. BRIEF DESCRIPTION OF THE DRAWINGS
[0009] Figure 1 A flow chart of a coal price forecasting method provided by an embodiment of the present invention; Figure 2 A schematic diagram of the structure of a coal price forecasting system provided by an embodiment of the present invention; Figure 3 A schematic diagram of the hardware structure of a possible electronic device provided by the present invention; Figure 4 A schematic diagram of the hardware structure of a possible computer-readable storage medium provided by the present invention. DETAILED DESCRIPTION
[0010] 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. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. In addition, the technical features in the various embodiments or single embodiments provided by the present invention can be arbitrarily combined with each other to form a feasible technical solution. This combination is not restricted by the sequence of steps and / or structural composition mode, but must be based on the ability of ordinary technicians in this field to implement it. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.
[0011] Figure 1 A flow chart of a coal price forecasting method provided by the present invention is as follows: Figure 1 As shown, the method includes: Step 1: Obtain historical prices of target coal to form an actual price time series of the target coal, screen the price time series for influencing factors, and the price time series and the screened influencing factors form an initial time series.
[0012] It is understood that in order to predict the future price of target coal, the embodiment of the present invention collects historical prices of target coal, and the historical actual prices of target coal at different times constitute the actual price time series of target coal. The influencing factors affecting the target coal price are screened to select the factors with relatively large impact on the target coal price.
[0013] In a possible implementation of the present invention, screening the price time series for influencing factors includes: Obtain historical coal prices and multiple factors that may affect coal prices; Calculating the correlation between each of the influencing factors and coal prices; The influencing factors whose correlation with the coal price is greater than a preset threshold are screened out.
[0014] Specifically, when screening factors influencing the target coal price, the correlation between each factor and the target coal price is calculated. Factors with correlations greater than a preset threshold are retained, and these factors are considered to have a certain impact on the target coal price. In this embodiment of the present invention, the screened factors primarily include unit power generation costs, electricity trends, and historical medium- and long-term coal prices. The target coal price time series and the corresponding influencing factors form an initial time series, which is subsequently used to train the prediction model.
[0015] Step 2: predicting the linear part of the initial time series based on the ARAM model to obtain a predicted value of the linear part.
[0016] It is understandable that the prediction model in the embodiment of the present invention includes an ARAM model and an LSTM model. The ARAM model is first used to predict the linear part in the initial time series to obtain a predicted value of the linear part.
[0017] In a possible implementation of the present invention, the linear portion of the initial time series is predicted based on the ARAM algorithm to obtain a predicted value of the linear portion, including: Perform stabilization on the initial time series and obtain the difference d and D values by the number of stabilization processes; Determine the values of p, q and P, Q of the ARAM model; The ARAM model is expressed as: ARAM=(p,d,q)×(P,D,Q) s Where p is the autoregressive order, q is the moving average order, d is the difference order, P, Q, D are the values of autoregressive, moving average and difference in the ARAM model respectively, and S is the cycle length period.
[0018] The initial time series is stabilized, and the values of the difference d and D are obtained by performing the stabilization process several times, including: Performing a single root test on the initial time series to determine whether the initial time series is stationary; If it is not stationary, the first-order difference is made using the difference method, and the sample autocorrelation function and the sample partial correlation function are calculated according to the sample value until the initial time series is stationary, and then the value of the order is determined to determine the value of the difference d and D; Determine the truncation status of the autocorrelation function and partial correlation function; The values of p, q, P and Q are determined according to the truncation status of the autocorrelation function and the partial correlation function.
[0019] Wherein, according to the truncation state of the autocorrelation function and the partial correlation function, the values of p, q, P and Q are determined, including: If the autocorrelation function is tailing and the partial autocorrelation function is truncated, the values of p and P are determined by the order of visualization; If the autocorrelation function is truncated and the partial autocorrelation function is smeared, determine the values of q and Q by visualizing the order; If the autocorrelation function and the partial autocorrelation function are both tailing, the values of p and P, q and Q can be determined by visualizing the order; The initial time series is decomposed, the value of the cycle length S is determined, and all parameters p, P, Q, q, d, D and S are obtained to construct the ARAM model.
[0020] After the ARAM model is constructed, the linear part of the time series is predicted based on the constructed ARAM.
[0021] Step 3: stripping the linear part from the initial time series to obtain the nonlinear part of the initial time series; Step 4: predict the nonlinear part based on the LSTM model to obtain a predicted value of the nonlinear part.
[0022] It can be understood that the linear part is separated from the initial time series to obtain the nonlinear part of the initial time series. The nonlinear part is predicted based on the LSTM model to obtain the predicted value of the nonlinear part.
[0023] In a possible implementation of the present invention, predicting the nonlinear part based on the LSTM model to obtain a predicted value of the nonlinear part includes: Dividing the nonlinear part of the initial time series into a training set and a test set, constructing the number of LSTM units in the LSTM model according to the training set, and training the LSTM model constructed based on the training set; The trained LSTM model is tested using a test set, and the LSTM model is optimized according to the test results until the prediction accuracy is achieved.
[0024] Step 5: Add the linear part prediction value and the nonlinear prediction value to obtain the price prediction time series value.
[0025] It can be understood that the linear part prediction value obtained by the ARAM model for the linear part of the initial time series and the nonlinear prediction of the LSTM model for the initial time series are added to obtain the price prediction time series value, which includes the coal price prediction value at each time.
[0026] Step 6: Compare the price prediction time series value with the actual price time series to obtain an error series.
[0027] Step 7: Optimize and adjust the ARAM model and the LSTM model based on the error sequence to obtain the optimized and adjusted ARAM model and the LSTM model.
[0028] It can be understood that the price prediction time series values obtained based on the ARAM model and the LSTM model are compared with the original target coal price time series values, the error sequence between the two is calculated, and the parameters of the ARAM model and the LSTM model are optimized and adjusted based on the error sequence to obtain the optimized and adjusted ARAM model and LSTM model.
[0029] Step 8: Based on the optimized and adjusted ARAM model and the LSTM model, the sequence of influencing factors to be predicted is predicted to obtain the coal price.
[0030] As you can see, through steps 6 and 7, we obtain the optimized and adjusted ARAM and LSTM models. The linear portion of the time series to be predicted is predicted using the ARAM model, while the nonlinear portion is predicted using the LSTM model. The predicted values for the linear and nonlinear portions are added together to obtain the coal price forecast.
[0031] See also Figure 2 , provides a coal price forecasting system according to an embodiment of the present invention, the system comprising: An acquisition module 201 is configured to acquire historical prices of target coal to form an actual price time series of the target coal, filter influencing factors of the price time series, and form an initial time series with the price time series and the filtered influencing factors; The first prediction module 202 is configured to predict the linear portion of the initial time series based on the ARAM model to obtain a linear portion prediction value; and to separate the linear portion from the initial time series to obtain a nonlinear portion of the initial time series, and to predict the nonlinear portion based on the LSTM model to obtain a nonlinear portion prediction value; and to add the linear portion prediction value and the nonlinear prediction value to obtain a price prediction time series value; A calculation module 203 is used to calculate the error sequence between the price prediction time series value and the actual price time series; An optimization module 204 is configured to optimize and adjust the ARAM model and the LSTM model based on the error sequence to obtain the optimized ARAM model and the LSTM model; The second prediction module 205 is used to predict the sequence of influencing factors to be predicted based on the optimized and adjusted ARAM model and the LSTM model to obtain the coal price.
[0032] It can be understood that the coal price prediction system provided by the present invention corresponds to the coal price prediction method provided in the aforementioned embodiments. The relevant technical features of the coal price prediction system can refer to the relevant technical features of the coal price prediction method, which will not be repeated here.
[0033] See also Figure 3 , Figure 3 Schematic diagram of an embodiment of an electronic device provided by an embodiment of the present invention. Figure 3 As shown, an embodiment of the present invention provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor 320. When the processor 320 executes the computer program 311, the steps of the coal price prediction method are implemented.
[0034] See also Figure 4 , Figure 4 Schematic diagram of an embodiment of a computer-readable storage medium provided by the present invention. Figure 4 As shown, this embodiment provides a computer-readable storage medium 400 on which a computer program 411 is stored. When the computer program 411 is executed by a processor, the steps of the coal price prediction method are implemented.
[0035] Embodiments of the present invention provide a coal price forecasting method, system, electronic device, and storage medium. The method obtains the historical price of target coal to form an actual price time series of the target coal, screens the price time series for influencing factors to form an initial time series, predicts the linear portion of the initial time series based on an ARAM model, predicts the nonlinear portion based on an LSTM model to obtain a predicted value for the nonlinear portion, compares the predicted price time series value with the actual price time series to obtain an error sequence, optimizes and adjusts the ARAM model and the LSTM model based on the error sequence, and predicts the sequence of influencing factors to be predicted based on the optimized and adjusted ARAM model and LSTM model to obtain the coal price. The present invention predicts the linear and nonlinear portions of the time series based on the ARAM model and the LSTM model, respectively, to achieve trend forecasting of coal prices.
[0036] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.
[0037] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0038] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0039] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.
[0040] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.
[0041] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.
[0042] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.
Claims
1. A coal price forecasting method, characterized in that: include: Obtaining historical prices of target coal to form an actual price time series of the target coal, screening influencing factors on the price time series, and using the price time series and the screened influencing factors to form an initial time series; Predicting the linear portion of the initial time series based on the ARAM model to obtain a predicted value of the linear portion; Stripping the linear portion from the initial time series to obtain a nonlinear portion of the initial time series; Predicting the nonlinear part based on the LSTM model to obtain a predicted value of the nonlinear part; Adding the linear part prediction value and the nonlinear prediction value to obtain a price prediction time series value; Comparing the price prediction time series value with the actual price time series to obtain an error series; Optimizing and adjusting the ARAM model and the LSTM model based on the error sequence to obtain the optimized and adjusted ARAM model and the LSTM model; Based on the optimized and adjusted ARAM model and the LSTM model, the sequence of influencing factors to be predicted is predicted to obtain the coal price.
2. The coal price forecasting method according to claim 1, characterized in that: The screening of influencing factors of the price time series includes: Obtain historical coal prices and multiple factors that may affect coal prices; Calculating the correlation between each of the influencing factors and coal prices; The influencing factors whose correlation with the coal price is greater than a preset threshold are screened out.
3. The coal price forecasting method according to claim 2, characterized in that: The screened influencing factors include unit power generation costs, electricity trends, and historical medium- and long-term coal prices.
4. The coal price forecasting method according to claim 1, characterized in that: The method of predicting the linear portion of the initial time series based on the ARAM algorithm to obtain a predicted value of the linear portion includes: Performing a stabilization process on the initial time series, and obtaining the values of the differences d and D according to the number of stabilization processes; Determining the values of p, q and P, Q of the ARAM model; The ARAM model is expressed as: ARAM=(p,d ,q)×(P,D ,Q) s Where p is the autoregressive order, q is the moving average order, d is the difference order, P, Q, D are the values of autoregressive, moving average and difference in the ARAM model respectively, and S is the cycle length period.
5. The coal price forecasting method according to claim 4, characterized in that: The initial time series is stabilized, and the values of the differences d and D are obtained by performing the stabilization process several times, including: Performing a single root test on the initial time series to determine whether the initial time series is stationary; If it is not stationary, the first-order difference is made using the difference method, and the sample autocorrelation function and the sample partial correlation function are calculated according to the sample value until the initial time series is stationary, and then the value of the order is determined to determine the value of the difference d and D; Determine the truncation status of the autocorrelation function and partial correlation function; The values of p, q, P and Q are determined according to the truncation status of the autocorrelation function and the partial correlation function.
6. The coal price forecasting method according to claim 5, characterized in that: Determining the values of p, q, P, and Q according to the truncation states of the autocorrelation function and the partial correlation function includes: If the autocorrelation function is tailing and the partial autocorrelation function is truncated, the values of p and P are determined by the order of visualization; If the autocorrelation function is truncated and the partial autocorrelation function is smeared, determine the values of q and Q by visualizing the order; If the autocorrelation function and the partial autocorrelation function are both tailing, the values of p and P, q and Q can be determined by visualizing the order; The initial time series is decomposed, the value of the cycle length S is determined, and all parameters p, P, Q, q, d, D and S are obtained to construct the ARAM model.
7. The coal price forecasting method according to claim 1, characterized in that: The predicting of the nonlinear part based on the LSTM model to obtain a predicted value of the nonlinear part includes: Dividing the nonlinear part of the initial time series into a training set and a test set, constructing the number of LSTM units in the LSTM model according to the training set, and training the LSTM model constructed based on the training set; The trained LSTM model is tested using a test set, and the LSTM model is optimized according to the test results until the prediction accuracy is achieved. Optimize until a high prediction accuracy is achieved.
8. A coal price forecasting system, characterized in that: include: An acquisition module is used to acquire historical prices of target coal to form an actual price time series of the target coal, screen influencing factors of the price time series, and the price time series and the screened influencing factors form an initial time series; a first prediction module, configured to predict the linear portion of the initial time series based on the ARAM model to obtain a linear portion prediction value; and to separate the linear portion from the initial time series to obtain a nonlinear portion of the initial time series, and to predict the nonlinear portion based on the LSTM model to obtain a nonlinear portion prediction value; and to add the linear portion prediction value and the nonlinear prediction value to obtain a price prediction time series value; A calculation module, used to calculate the error sequence between the price prediction time series value and the actual price time series; An optimization module, configured to optimize and adjust the ARAM model and the LSTM model based on the error sequence to obtain the optimized ARAM model and the LSTM model; The second prediction module is used to predict the sequence of influencing factors to be predicted based on the optimized and adjusted ARAM model and the LSTM model to obtain the coal price.
9. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the processor is used to implement the steps of the coal price forecasting method according to any one of claims 1 to 7 when executing a computer management program stored in the memory.
10. A computer-readable storage medium, characterized in that A computer management program is stored thereon, and when the computer management program is executed by the processor, the steps of the coal price forecasting method according to any one of claims 1 to 7 are implemented.