A power price early warning method, device, storage medium and system

Abstract

The application discloses a power price early warning method and device, a storage medium and a system. Through smoothness verification and co-integration verification based on historical data of gas prices and electricity prices, co-integration verification prediction results are calculated after both verifications pass. Then, the relationship between the current electricity price and the historical gas price and the historical electricity price is quantitatively studied based on a VAR model, prediction is performed to obtain model prediction results, early warning judgment is performed by comparing the electricity price prediction errors of the co-integration equation and the VAR model expression, the method is easy to operate and the result is intuitive, the influence of gas price fluctuation on electricity price prediction is fully considered, the early warning method, the device, the storage medium and the system improve the early warning accuracy of the power price, thereby providing data support for improving the dispatching efficiency.

Description

Technical Field

[0001] This invention relates to the field of electricity price early warning technology, and in particular to an electricity price early warning method, device, computer-readable storage medium and system. Background Technology

[0002] Power grid dispatch management refers to the management of power grid production and operation, power grid dispatching systems, and personnel duties by power grid dispatching agencies in accordance with relevant regulations to ensure the safe, high-quality, and economical operation of the power grid. It generally includes dispatch operation management, dispatch planning management, relay protection and automatic safety device management, power grid dispatch automation management, power communication management, hydropower plant and reservoir dispatch management, and power system personnel training management. The natural gas market plays a crucial role in promoting energy market development against the backdrop of the current global energy shift towards low-carbon and zero-carbon development, and the natural gas sector will have enormous development potential in the future. However, the frequent price risk events in the global natural gas and electricity markets in recent years have made countries realize the significant importance of studying the linkage mechanism between natural gas and electricity prices, and the importance of electricity price early warning research considering the impact of gas price fluctuations. The linkage between gas and electricity prices can reflect the stable internal mechanism between the gas and electricity markets, thus providing a basis for judgment in electricity price early warning research.

[0003] In existing technologies, the power resources of the power grid are usually dispatched based on electricity prices.

[0004] However, existing technologies still have the following drawbacks: the factors considered in electricity price early warning are relatively singular, and the linkage with natural gas prices is not taken into account, resulting in poor accuracy in electricity price prediction and consequently poor dispatch efficiency.

[0005] Therefore, there is a current need for an early warning method, device, computer-readable storage medium, and system for electricity prices to overcome the aforementioned deficiencies in the prior art. Summary of the Invention

[0006] This invention provides a method, apparatus, computer-readable storage medium, and system for early warning of electricity prices, thereby improving the accuracy of early warning of electricity prices and providing data support for improving dispatch efficiency.

[0007] An embodiment of the present invention provides an early warning method for electricity prices. The early warning method includes: acquiring electricity market price data and natural gas market price data; performing a stationarity check on the electricity market price data and the natural gas market price data; if the stationarity check passes, performing a cointegration check on the electricity market price data and the natural gas market price data, and obtaining a cointegration check prediction result; if the cointegration check passes, constructing a gas-electricity vector autoregressive model based on the electricity market price data and the natural gas market price data, and calculating the model prediction result based on the gas-electricity vector autoregressive model, the electricity market price data, and the natural gas market price data; calculating the price fluctuation of electricity prices based on a preset price fluctuation calculation formula, the cointegration check prediction result, and the model prediction result; and issuing an electricity price early warning when the price fluctuation exceeds a preset fluctuation threshold.

[0008] As an improvement to the above scheme, the early warning method further includes: performing a Granger causality test on gas prices and electricity prices based on the electricity market price data and the natural gas market price data, and determining whether the gas prices and electricity prices influence each other based on the test results; when it is determined that the gas prices and electricity prices influence each other, performing a pulse response dynamic analysis on the gas prices and electricity prices based on the gas-electricity vector autoregression model, the electricity market price data, and the natural gas market price data, and determining whether the gas prices and electricity prices will have a dynamic impact on the electricity prices and gas prices when pulse fluctuations occur; if so, using a preset contribution analysis method, quantifying the impact of gas prices and electricity prices on the power system to obtain the gas price contribution and electricity price contribution, and analyzing the dynamic changes in gas and electricity prices based on the gas price contribution and electricity price contribution.

[0009] As an improvement to the above scheme, the stationarity of the electricity market price data and the natural gas market price data is verified, specifically including: verifying the stationarity of electricity prices and gas prices respectively according to the preset electricity price regression equation, the preset gas price regression equation and the unit root test equation.

[0010] As an improvement to the above scheme, cointegration verification is performed on the electricity market price data and the natural gas market price data to obtain cointegration verification prediction results. Specifically, this includes: fitting the electricity market price data and the natural gas market price data to obtain a gas-electricity function relationship; calculating the optimal parameter set in the gas-electricity function relationship using the least squares method according to a preset formula for the sum of squared prediction errors; and substituting the optimal parameter set into the gas-electricity function relationship to obtain the cointegration verification prediction results.

[0011] As an improvement to the above scheme, the specific model expression of the gas-electric vector autoregression model is as follows: ε t =[ε 1t ,ε 2t ] T , φ0=[φ 10 ,φ 20 ] T , In the formula, p is the lag order in the VAR model of gas and electricity prices; T is the total number of days of collected price data; φ0 is a column vector consisting of constant terms in the respective regression equations of electricity and gas prices; φ i This is the coefficient matrix between current and lagged data for gas and electricity prices; ε t The perturbation terms of the model are column vectors, each independent and uncorrelated with a mean of 0.

[0012] As an improvement to the above scheme, the price fluctuation calculation formula is as follows: Among them, Y t1 Y is the electricity price obtained through the cointegration equation; t2 The electricity price is obtained through the VAR expression.

[0013] As an improvement to the above scheme, after performing stationarity verification on the electricity market price data and the natural gas market price data, the early warning method further includes: when the stationarity verification fails, performing differential processing on the natural gas market price data and the electricity market price data respectively to obtain gas price differential sequences and electricity price differential sequences; performing stationarity verification on the gas price differential sequences and the electricity price differential sequences; if the gas price differential sequences and the electricity price differential sequences are integrated of the same order, then performing cointegration verification on the gas price differential sequences and the electricity price differential sequences to obtain cointegration verification prediction results; when the cointegration verification passes, constructing a gas-electricity vector autoregressive model based on the gas price differential sequences and the electricity price differential sequences, and calculating the model prediction results based on the gas-electricity vector autoregressive model, the gas price differential sequences, and the electricity price differential sequences; calculating the price fluctuation of electricity prices based on a preset price fluctuation calculation formula, the cointegration verification prediction results, and the model prediction results, and issuing an electricity price early warning when the price fluctuation exceeds a preset fluctuation threshold.

[0014] Another embodiment of the present invention provides an early warning device for electricity prices. The early warning device includes a cointegration prediction unit, a model prediction unit, and a fluctuation early warning unit. The cointegration prediction unit acquires electricity market price data and natural gas market price data, performs stationarity checks on the electricity market price data and the natural gas market price data, and performs cointegration checks on the electricity market price data and the natural gas market price data when the stationarity check passes, obtaining a cointegration check prediction result. The model prediction unit, when the cointegration check passes, constructs a gas-electricity vector autoregressive model based on the electricity market price data and the natural gas market price data, and calculates the model prediction result based on the gas-electricity vector autoregressive model, the electricity market price data, and the natural gas market price data. The fluctuation early warning unit calculates the price fluctuation of electricity prices based on a preset price fluctuation calculation formula, the cointegration check prediction result, and the model prediction result, and issues an electricity price early warning when the price fluctuation exceeds a preset fluctuation threshold.

[0015] As an improvement to the above scheme, the early warning device further includes a dynamic analysis unit, which is used to: perform Granger causality tests on gas prices and electricity prices based on the electricity market price data and the natural gas market price data, and determine whether the gas prices and electricity prices affect each other based on the test results; when it is determined that the gas prices and electricity prices affect each other, perform impulse response dynamic analysis on the gas prices and electricity prices based on the gas-electricity vector autoregression model, the electricity market price data, and the natural gas market price data, and determine whether the gas prices and electricity prices will have a corresponding dynamic impact on the electricity prices and gas prices when impulse fluctuations occur; if so, quantify the impact of gas prices and electricity prices on the power system using a preset contribution analysis method to obtain the gas price contribution degree and the electricity price contribution degree, and analyze the dynamic changes in gas and electricity prices based on the gas price contribution degree and the electricity price contribution degree.

[0016] As an improvement to the above scheme, the early warning device further includes a secondary verification unit, which is used to: when the stationarity verification fails, perform differential processing on the natural gas market price data and the electricity market price data respectively to obtain the gas price differential sequence and the electricity price differential sequence; perform stationarity verification on the gas price differential sequence and the electricity price differential sequence; if the gas price differential sequence and the electricity price differential sequence are integrated of the same order, then perform cointegration verification on the gas price differential sequence and the electricity price differential sequence to obtain the cointegration verification prediction result; when the cointegration verification passes, construct a gas-electricity vector autoregressive model based on the gas price differential sequence and the electricity price differential sequence, and calculate the model prediction result based on the gas-electricity vector autoregressive model, the gas price differential sequence and the electricity price differential sequence; calculate the price fluctuation of electricity price based on the preset price fluctuation calculation formula, the cointegration verification prediction result and the model prediction result, and issue an electricity price early warning when the price fluctuation exceeds a preset fluctuation threshold.

[0017] As an improvement to the above scheme, the cointegration prediction unit is further used to: fit the electricity market price data and the natural gas market price data to obtain the gas-electricity function relationship; calculate the optimal parameter set in the gas-electricity function relationship using the least squares method according to the preset prediction error sum of squares formula; and substitute the optimal parameter set into the gas-electricity function relationship to obtain the cointegration verification prediction result.

[0018] As an improvement to the above scheme, the cointegration prediction unit is also used to: perform stationarity checks on electricity prices and gas prices respectively based on preset electricity price regression equations, preset gas price regression equations, and unit root test equations.

[0019] Another embodiment of the present invention provides a computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the electricity price early warning method as described above.

[0020] Another embodiment of the present invention provides an early warning system for electricity prices, the early warning system including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the electricity price early warning method as described above.

[0021] Compared with existing technologies, this technical solution has the following beneficial effects:

[0022] This invention provides a method, device, storage medium, and system for early warning of electricity prices. It performs stationarity and cointegration checks based on historical gas and electricity price data. After both checks pass, the cointegration check prediction result is calculated. Subsequently, based on a VAR model, the relationship between the current electricity price and historical gas and electricity prices is quantitatively studied, and predictions are made to obtain the model prediction results. Early warning judgment is made by comparing the electricity price prediction errors of the cointegration equation and the VAR model expression. The method is easy to operate, provides intuitive results, and fully considers the impact of gas price fluctuations on electricity price prediction. This early warning method, device, storage medium, and system improve the accuracy of electricity price early warning, thereby providing data support for improving dispatch efficiency. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating an early warning method for electricity prices according to an embodiment of the present invention;

[0024] Figure 2 This is a schematic diagram of the structure of an early warning device for electricity prices provided in an embodiment of the present invention. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Specific Implementation Example 1

[0027] The present invention first describes a method for early warning of electricity prices. Figure 1 This is a flowchart illustrating an early warning method for electricity prices provided in an embodiment of the present invention.

[0028] like Figure 1 As shown, the early warning method includes:

[0029] S1: Obtain electricity market price data and natural gas market price data, perform stationarity verification on the electricity market price data and natural gas market price data, and if the stationarity verification is passed, perform cointegration verification on the electricity market price data and natural gas market price data, and obtain the cointegration verification prediction result.

[0030] Considering the potential for multicollinearity and spurious regression due to heteroscedasticity in gas and electricity price time series, it is generally advisable to first take the logarithm of the original data and then process it. Let the processed natural gas and electricity price time series be X, respectively. t and Y tUsing electricity price time series Y t Taking the stationarity test as an example, consider the electricity price regression equation:

[0031] Y t =kY t-1 +ε t ;

[0032] In the formula, k is the regression coefficient, and ε t It is the error term and follows an independent and identically distributed pattern.

[0033] The electricity price at time t is represented using the data from the previous n days:

[0034]

[0035] If k=1, meaning the electricity price time series has a unit root, the variance of the electricity price time series will continuously increase, the influence of the residuals cannot be eliminated, and the series is unstable. The Augmented Dickey-Fuller (ADF) test determines whether the time series is stable based on the existence of a unit root. First, the Ordinary Least Squares (OLS) method is used to estimate... We then construct the test statistic t to determine whether the equation contains at least one unit root:

[0036]

[0037]

[0038] Based on the probability results obtained from t(k), it can be determined whether the electricity price time series is stable; similarly, for the gas price time series X... t The same stationarity test should also be performed.

[0039] In one embodiment, the stationarity verification of the electricity market price data and the natural gas market price data specifically includes: performing stationarity verification on the electricity price and the gas price respectively according to the preset electricity price regression equation, the preset gas price regression equation and the unit root test equation.

[0040] After passing the stationarity test, a cointegration test should be conducted on the long-term equilibrium relationship between gas and electricity prices. The specific implementation method is as follows:

[0041] When a stable intrinsic mechanism exists between gas and electricity prices, a stable functional relationship between them can be obtained through fitting and regression using long-term data; this is called a long-term equilibrium relationship. Even if gas or electricity prices experience short-term fluctuations, they can still maintain a static and stable equilibrium state. To study the long-term static relationship between gas and electricity prices, the following analysis examines gas and electricity price series under different stability states, yielding the gas-electricity functional relationship:

[0042] Y t =aX t +b;

[0043] If the gas price X t With electricity price Y t The stationarity test demonstrates that the relationship and characteristics between gas and electricity prices, as studied using historical data, will remain unchanged over a certain period of time, thus possessing practical significance and ensuring that spurious regression will not occur between gas and electricity prices. Therefore, a fitting regression analysis can be directly performed.

[0044] The optimal parameters in the regression equations for gas and electricity prices are estimated using the OLS method. and The regression equation is:

[0045]

[0046] In the formula, This indicates an estimated electricity price.

[0047] According to the OLS principle, when the optimal parameters are selected, the actual electricity price data Y t and estimated electricity price The sum of squares of the differences is minimized, Q:

[0048]

[0049] Solving for:

[0050]

[0051] However, in practical applications, most of the original time series obtained do not meet the strict stationarity requirement. Therefore, it is necessary to further test the original series by differentiating them. If the original series are integrated of the same order, then the stationarity requirement is met.

[0052] Perform first-order differences on gas and electricity prices and then perform stationarity checks again:

[0053]

[0054] If the time series is stationary when the difference is made for the dth time, it is said to be integrated of order d. Therefore, only when the gas price and electricity price series are integrated of the same or lower order can they meet the stationarity requirement. Subsequent studies will assume that the series meet the stationarity test and further determine whether there is a cointegration relationship between them.

[0055] Repeat the aforementioned fitting regression steps; the regression equation at this point is also called the cointegration equation, with respect to the electricity price residual term. Perform a unit root test to determine if the cointegration expression is correct. If the residual term passes the unit root test, it indicates that there is a long-term equilibrium relationship between gas prices and electricity prices, and the cointegration expression is valid.

[0056] In one embodiment, cointegration verification is performed on the electricity market price data and the natural gas market price data to obtain cointegration verification prediction results. Specifically, this includes: fitting the electricity market price data and the natural gas market price data to obtain a gas-electricity function relationship; calculating the optimal parameter set in the gas-electricity function relationship using the least squares method according to a preset formula for the sum of squared prediction errors; and substituting the optimal parameter set into the gas-electricity function relationship to obtain the cointegration verification prediction results.

[0057] In one embodiment, after performing stationarity checks on the electricity market price data and the natural gas market price data, the early warning method further includes: when the stationarity check fails, performing differential processing on the natural gas market price data and the electricity market price data respectively to obtain gas price differential sequences and electricity price differential sequences; performing stationarity checks on the gas price differential sequences and the electricity price differential sequences; if the gas price differential sequences and the electricity price differential sequences are integrated of the same order, then performing cointegration checks on the gas price differential sequences and the electricity price differential sequences to obtain cointegration check prediction results; when the cointegration check passes, constructing a gas-electricity vector autoregressive model based on the gas price differential sequences and the electricity price differential sequences, and calculating the model prediction results based on the gas-electricity vector autoregressive model, the gas price differential sequences, and the electricity price differential sequences; calculating the price fluctuation of electricity prices based on a preset price fluctuation calculation formula, the cointegration check prediction results, and the model prediction results, and issuing an electricity price early warning when the price fluctuation exceeds a preset fluctuation threshold.

[0058] S2: When the cointegration verification passes, construct a gas-electricity vector autoregressive model based on the electricity market price data and the natural gas market price data, and calculate the model prediction results based on the gas-electricity vector autoregressive model, the electricity market price data, and the natural gas market price data.

[0059] In one embodiment, the model expression of the gas-electric vector autoregressive model is specifically as follows:

[0060]

[0061]

[0062] In the formula, p is the lag order in the VAR model of gas and electricity prices; T is the total number of days of collected price data; φ0 is the column vector composed of constant terms in the respective regression equations of electricity and gas prices; φ i This is the coefficient matrix between current and lagged data for gas and electricity prices; ε t The perturbation terms of the model are column vectors, each independent and uncorrelated with a mean of 0.

[0063] In one embodiment, to ensure a linkage between gas and electricity prices, thus avoiding calculations when they are not linked, further dynamic analysis of gas-electricity prices based on a VAR model is required. The specific implementation method is as follows:

[0064] 1) Perform Granger causality tests on gas-electricity prices. This involves analyzing a first-order lag model: "Gas price X..." t Is it the cause of electricity price Y? t Taking the reason for the change as an example, the null hypothesis at this time is "gas price X". t Not electricity price Y t The reason for the change

[0065] (1) Regarding electricity price Y t The unconstrained regression model (u) and the constrained regression model (r) are estimated.

[0066] Among them, the unconstrained regression model (u):

[0067] Y t =φ 11 (1)Y t-1 +φ 12 (1)X t-1 +ε 1t ;

[0068] Constrained regression model (r):

[0069] Y t =kY t-1 +ε 1t ;

[0070] The estimated coefficients were obtained using the VAR model and OLS, respectively. and Calculate the sum of squared residuals of the model and construct the F-statistic:

[0071]

[0072]

[0073]

[0074] In the formula, F represents the variance statistic, and k = 2p, i.e., p = 1 and k = 2.

[0075] Based on the statistical results, the null hypothesis "gas price X" can be redefined. t Not electricity price Y t The reason for the change is determined by "F". <F α (p,nk), then it is considered that under the premise of a significance level of α, "gas price X" t Not electricity price Y t "The reason for the change"; otherwise, reject the null hypothesis.

[0076] (2) Change gas price X t With electricity price Y t The causal order is determined, and the hypothesis "electricity price Y" is tested using the same method as in (1). t It did not cause gas prices X t The reasons for the change.

[0077] (3) If the test result also rejects "gas price X" t It does not cause electricity price Y t The reasons for the change and the acceptance of "electricity price Y" t It did not cause gas prices X t "The reasons for the change" leads to the conclusion that "gas price is a Granger cause of electricity price".

[0078] 2) Dynamic pulse response to gas-electricity prices. Taking a first-order lag model as an example, the VAR model is as follows:

[0079]

[0080] Let X -1 =X -2 =Y -1 =Y -2 =0, when testing the impact of sudden changes in natural gas prices on the model, it is necessary to adjust the gas price X. t Perform the assignment, let ε 10 =1,ε 20 =0,ε 1t =ε 2t =0 (t=1,2,…,T), and X1,X2,…,X can be calculated according to the model. T With Y1, Y2, ..., Y TThis allows us to obtain the dynamic changes in gas and electricity prices following pulse fluctuations in natural gas prices. Similarly, we can examine the impact of sudden changes in electricity prices on the system.

[0081] 3) Variance decomposition analysis was performed on gas and electricity prices. Variance decomposition further quantified the impact of shocks on gas and electricity prices and described the contribution of different shocks to the fluctuations in gas and electricity prices.

[0082] Consider the two-variable, first-order sequence form of gas and electricity prices with respect to random disturbance terms:

[0083]

[0084] In the formula a 11 a 12 a 21 a 22 The corresponding coefficients of the random disturbance term

[0085] To determine the contribution of each disturbance term to the system variance, the Rate of Variance Contribution (RVC) is defined. If the disturbance terms are pairwise uncorrelated, then:

[0086]

[0087]

[0088]

[0089] In the formula, σ represents the variance of the sequence, RVC Y and RVC X These represent the relative contributions of electricity price and natural gas price disturbances to the overall system variance, respectively.

[0090] Dynamic analysis of gas-electricity prices can provide electricity market participants with more comprehensive price correlation information, provide more sufficient basis for judgment on electricity price early warning, and help judge the severity of future electricity price fluctuations.

[0091] In one embodiment, the early warning method further includes: performing a Granger causality test on gas prices and electricity prices based on the electricity market price data and the natural gas market price data, and determining whether the gas prices and electricity prices influence each other based on the test results; when it is determined that the gas prices and electricity prices influence each other, performing a pulse response dynamic analysis on the gas prices and electricity prices based on the gas-electricity vector autoregression model, the electricity market price data, and the natural gas market price data, and determining whether the gas prices and electricity prices will have a dynamic impact on the electricity prices and gas prices when pulse fluctuations occur; if so, using a preset contribution analysis method, quantifying the impact of gas prices and electricity prices on the power system to obtain the gas price contribution degree and the electricity price contribution degree, and analyzing the dynamic changes in gas and electricity prices based on the gas price contribution degree and the electricity price contribution degree.

[0092] S3: Calculate the price fluctuation of electricity based on the preset price fluctuation calculation formula, the cointegration verification prediction result, and the model prediction result, and issue an electricity price warning when the price fluctuation exceeds the preset fluctuation threshold.

[0093] Since the cointegration equation for gas-electricity prices is based on a long-term stable price function relationship derived from historical data, reflecting the inherent mechanism of current gas-electricity prices, the electricity price Y obtained through the cointegration equation... t1 It can serve as a benchmark for judging future electricity prices. The VAR model expression for gas-electricity prices takes into account the time lag phenomenon of price time series, and mainly provides a quantitative explanation of the relationship between current electricity prices and historical electricity and gas prices from a short-term perspective. Therefore, the electricity price Y obtained through the VAR expression... t2 The predictions for future electricity prices are closer to reality. When Y t2 and Y t1 The prediction error exceeds the threshold θ, i.e. The system will issue an electricity price warning at that time.

[0094] In one embodiment, the price fluctuation calculation formula is:

[0095]

[0096] In the formula, Y t1 Y is the electricity price obtained through the cointegration equation; t2 Let θ be the electricity price obtained through the VAR expression. In one embodiment, θ is 5%.

[0097] This invention describes an early warning method for electricity prices. It performs stationarity and cointegration checks based on historical gas and electricity price data. After both checks pass, the cointegration check prediction result is calculated. Subsequently, based on a VAR model, the relationship between the current electricity price and historical gas and electricity prices is quantitatively studied, and predictions are made to obtain the model prediction results. Early warning judgments are made by comparing the electricity price prediction errors of the cointegration equation and the VAR model expression. This method is easy to operate, provides intuitive results, and fully considers the impact of gas price fluctuations on electricity price prediction. This early warning method improves the accuracy of electricity price early warnings, thereby providing data support for improving dispatch efficiency. Specific Implementation Example 2

[0099] In addition to the methods described above, embodiments of the present invention also disclose an early warning device for electricity prices. Figure 2 This is a schematic diagram of the structure of an early warning device for electricity prices provided in an embodiment of the present invention.

[0100] like Figure 2 As shown, the early warning device includes a cointegration prediction unit 11, a model prediction unit 12, and a fluctuation early warning unit 13.

[0101] The cointegration prediction unit 11 is used to acquire electricity market price data and natural gas market price data, perform stationarity verification on the electricity market price data and natural gas market price data, and perform cointegration verification on the electricity market price data and natural gas market price data when the stationarity verification is passed, and obtain the cointegration verification prediction result.

[0102] In one embodiment, the cointegration prediction unit is further configured to: fit the electricity market price data and the natural gas market price data to obtain a gas-electricity function relationship; calculate the optimal parameter set in the gas-electricity function relationship using the least squares method according to a preset formula for the sum of squared prediction errors; and substitute the optimal parameter set into the gas-electricity function relationship to obtain the cointegration verification prediction result.

[0103] In one embodiment, the cointegration prediction unit is further configured to: perform stationarity checks on electricity prices and gas prices respectively based on preset electricity price regression equations, preset gas price regression equations, and unit root test equations.

[0104] The model prediction unit 12 is used to construct a gas-electricity vector autoregressive model based on the electricity market price data and the natural gas market price data when the cointegration verification is passed, and to calculate the model prediction results based on the gas-electricity vector autoregressive model, the electricity market price data and the natural gas market price data.

[0105] The fluctuation warning unit 13 is used to calculate the price fluctuation of electricity based on the preset price fluctuation calculation formula, the cointegration verification prediction result and the model prediction result, and to issue an electricity price warning when the price fluctuation exceeds the preset fluctuation threshold.

[0106] In one embodiment, the early warning device further includes a dynamic analysis unit, which is used to: perform Granger causality tests on gas prices and electricity prices based on the electricity market price data and the natural gas market price data, and determine whether the gas prices and electricity prices influence each other based on the test results; when it is determined that the gas prices and electricity prices influence each other, perform impulse response dynamic analysis on the gas prices and electricity prices based on the gas-electricity vector autoregression model, the electricity market price data, and the natural gas market price data, and determine whether the gas prices and electricity prices will have a dynamic impact on the electricity prices and gas prices when impulse fluctuations occur; if so, quantify the impact of gas prices and electricity prices on the power system using a preset contribution analysis method to obtain the gas price contribution degree and the electricity price contribution degree, and analyze the dynamic changes in gas and electricity prices based on the gas price contribution degree and the electricity price contribution degree.

[0107] In one embodiment, the early warning device further includes a secondary verification unit, which is configured to: when the stationarity verification fails, perform differential processing on the natural gas market price data and the electricity market price data respectively to obtain gas price differential sequences and electricity price differential sequences; perform stationarity verification on the gas price differential sequences and the electricity price differential sequences; if the gas price differential sequences and the electricity price differential sequences are integrated of the same order, perform cointegration verification on the gas price differential sequences and the electricity price differential sequences to obtain cointegration verification prediction results; when the cointegration verification passes, construct a gas-electricity vector autoregressive model based on the gas price differential sequences and the electricity price differential sequences, and calculate the model prediction results based on the gas-electricity vector autoregressive model, the gas price differential sequences, and the electricity price differential sequences; calculate the price fluctuation of electricity prices based on a preset price fluctuation calculation formula, the cointegration verification prediction results, and the model prediction results, and issue an electricity price early warning when the price fluctuation exceeds a preset fluctuation threshold.

[0108] If the unit integrated into the early warning device is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. That is, another embodiment of the present invention provides a computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the electricity price early warning method as described above.

[0109] The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0110] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between units indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0111] This invention describes an early warning device and storage medium for electricity prices. It performs stationarity and cointegration checks based on historical gas and electricity price data. After both checks pass, the cointegration check prediction result is calculated. Subsequently, based on a VAR model, the relationship between the current electricity price and historical gas and electricity prices is quantitatively studied, and predictions are made to obtain the model prediction results. Early warning judgments are made by comparing the electricity price prediction errors of the cointegration equation and the VAR model expression. The device is easy to operate, provides intuitive results, and fully considers the impact of gas price fluctuations on electricity price prediction. This early warning device and storage medium improve the accuracy of electricity price early warnings, thereby providing data support for improving dispatch efficiency. Specific Implementation Example 3

[0113] In addition to the methods and apparatus described above, embodiments of the present invention also describe an early warning system for electricity prices.

[0114] The early warning system includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the electricity price early warning method as described above.

[0115] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the device, connecting various parts of the device via various interfaces and lines.

[0116] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0117] This invention describes an early warning system for electricity prices. It performs stationarity and cointegration checks based on historical gas and electricity price data. After both checks pass, the cointegration check prediction result is calculated. Subsequently, based on a VAR model, the relationship between the current electricity price and historical gas and electricity prices is quantitatively studied, and predictions are made to obtain the model prediction results. Early warning judgments are made by comparing the electricity price prediction errors of the cointegration equation and the VAR model expression. The system is easy to operate, provides intuitive results, and fully considers the impact of gas price fluctuations on electricity price prediction. This early warning system improves the accuracy of electricity price early warnings, thereby providing data support for improving dispatch efficiency.

[0118] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for early warning of electricity prices, characterized in that, The early warning method includes: Acquire electricity market price data and natural gas market price data, perform stationarity verification on the electricity market price data and natural gas market price data, and when the stationarity verification is passed, perform cointegration verification on the electricity market price data and natural gas market price data, and obtain the cointegration verification prediction results; When the cointegration verification is passed, a gas-electricity vector autoregressive model is constructed based on the electricity market price data and the natural gas market price data, and the model prediction results are calculated based on the gas-electricity vector autoregressive model, the electricity market price data, and the natural gas market price data. Based on the preset price fluctuation calculation formula, the cointegration verification prediction results, and the model prediction results, the price fluctuation of electricity is calculated, and an electricity price warning is issued when the price fluctuation exceeds a preset fluctuation threshold. The early warning method further includes, after performing stability checks on the electricity market price data and the natural gas market price data: When the stationarity test fails, the natural gas market price data and the electricity market price data are differentially processed to obtain the gas price differential sequence and the electricity price differential sequence respectively. The gas price difference sequence and the electricity price difference sequence are subjected to stationarity verification. If the gas price difference sequence and the electricity price difference sequence are integrated of the same order, then the gas price difference sequence and the electricity price difference sequence are subjected to cointegration verification to obtain the cointegration verification prediction result. When the cointegration verification passes, a gas-electricity vector autoregressive model is constructed based on the gas price difference sequence and the electricity price difference sequence, and the model prediction result is calculated based on the gas-electricity vector autoregressive model, the gas price difference sequence, and the electricity price difference sequence. Based on the preset price fluctuation calculation formula, the cointegration verification prediction results, and the model prediction results, the price fluctuation of electricity is calculated, and an electricity price warning is issued when the price fluctuation exceeds a preset fluctuation threshold.

2. The method for early warning of electricity prices according to claim 1, characterized in that, The early warning method also includes: Based on the electricity market price data and the natural gas market price data, a Granger causality test is performed on the gas price and the electricity price, and based on the test results, it is determined whether the gas price and the electricity price affect each other. When it is determined that the gas price and the electricity price influence each other, the gas price and the electricity price are subjected to impulse response dynamic analysis based on the gas-electricity vector autoregression model, the electricity market price data, and the natural gas market price data. The analysis determines whether the gas price and the electricity price will have a dynamic impact on the electricity price and the gas price when impulse fluctuations occur. If so, the impact of gas prices and electricity prices on the power system is quantitatively analyzed using a preset contribution analysis method to obtain the contribution of gas prices and the contribution of electricity prices. Based on the contribution of gas prices and the contribution of electricity prices, the dynamic changes in gas and electricity prices are analyzed.

3. The method for early warning of electricity prices according to claim 2, characterized in that, The stability of the electricity market price data and the natural gas market price data is verified, specifically including: Based on the preset regression equations for electricity prices, gas prices, and unit root test equations, the stationarity of electricity and gas prices is verified.

4. The method for early warning of electricity prices according to claim 3, characterized in that, Cointegration verification is performed on the electricity market price data and the natural gas market price data to obtain the cointegration verification prediction results, specifically including: The electricity market price data and the natural gas market price data are fitted to obtain the gas-electricity functional relationship. The optimal parameter set in the gas-electric function relationship is calculated using the least squares method according to the preset formula for the sum of squared prediction errors. Substitute the optimal parameter set into the gas-electric function relationship to obtain the cointegration verification prediction result.

5. The method for early warning of electricity prices according to claim 4, characterized in that, The specific model expression of the gas-electric vector autoregressive model is as follows: ； ， ， ， ； In the formula, This represents the lag order in the VAR model for gas and electricity prices. The total number of days for which price data was collected; It is a column vector consisting of the constant terms in the respective regression equations of electricity price and gas price; This is the coefficient matrix between current data and lagged data for gas and electricity prices; The perturbation terms of the model are column vectors, each independent and uncorrelated with a mean of 0.

6. The method for early warning of electricity prices according to claim 5, characterized in that, The formula for calculating price fluctuations is: ； in, The electricity price is obtained through the cointegration equation; The electricity price is obtained through the VAR expression.

7. An early warning device for electricity prices, characterized in that, The early warning device includes a cointegration prediction unit, a model prediction unit, and a fluctuation early warning unit, wherein... The cointegration prediction unit is used to acquire electricity market price data and natural gas market price data, perform stationarity verification on the electricity market price data and natural gas market price data, and when the stationarity verification is passed, perform cointegration verification on the electricity market price data and natural gas market price data to obtain the cointegration verification prediction result. The model prediction unit is used to construct a gas-electricity vector autoregressive model based on the electricity market price data and the natural gas market price data when the cointegration verification passes, and to calculate the model prediction results based on the gas-electricity vector autoregressive model, the electricity market price data, and the natural gas market price data. The fluctuation early warning unit is used to calculate the price fluctuation of electricity prices based on the preset price fluctuation calculation formula, the cointegration verification prediction result and the model prediction result, and to issue an electricity price early warning when the price fluctuation exceeds the preset fluctuation threshold. The early warning method further includes, after performing stability checks on the electricity market price data and the natural gas market price data: When the stationarity test fails, the natural gas market price data and the electricity market price data are differentially processed to obtain the gas price differential sequence and the electricity price differential sequence respectively. The gas price difference sequence and the electricity price difference sequence are subjected to stationarity verification. If the gas price difference sequence and the electricity price difference sequence are integrated of the same order, then the gas price difference sequence and the electricity price difference sequence are subjected to cointegration verification to obtain the cointegration verification prediction result. When the cointegration verification passes, a gas-electricity vector autoregressive model is constructed based on the gas price difference sequence and the electricity price difference sequence, and the model prediction result is calculated based on the gas-electricity vector autoregressive model, the gas price difference sequence, and the electricity price difference sequence. Based on the preset price fluctuation calculation formula, the cointegration verification prediction results, and the model prediction results, the price fluctuation of electricity is calculated, and an electricity price warning is issued when the price fluctuation exceeds a preset fluctuation threshold.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device containing the computer-readable storage medium to perform the electricity price early warning method as described in any one of claims 1 to 6.

9. An early warning system for electricity prices, characterized in that, The early warning system includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the electricity price early warning method as described in any one of claims 1 to 6.

Images