Power, carbon and fossil fuel scheduling method and system based on co-integration theory

Abstract

The invention discloses an electric power, carbon and fossil fuel scheduling method and system based on a co-integration theory. The method comprises the following steps: acquiring a market price sequence in an electric power transaction period; based on a co-integration theory, co-integration vectors among the market price sequences are calculated; establishing a comprehensive power generation cost index model according to the market price sequence; and according to the co-integration vector, optimizing the comprehensive power generation cost index model, and generating a scheduling scheme of the power material. By adopting the embodiment of the invention, the long-term equilibrium relationship and the short-term dynamic adjustment characteristics among the prices of the electric power, the carbon and the fossil fuel can be analyzed based on the co-integration theory, the original information can be reserved as much as possible, the relevance among markets can be accurately captured, and the market competitiveness can be improved. And electric power materials such as carbon and fossil fuel are optimally dispatched according to a co-integration analysis result, so that the economic and environmental values can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of market price correlation calculation, and in particular to a method and system for dispatching electricity, carbon and fossil fuels based on cointegration theory. Background Art

[0002] The existing methods for studying the price correlation of multiple markets can be roughly divided into three categories. The first category is to control and change the key influencing factors of the power market and the power material market by constructing a system dynamics model, simulating the dynamic behavior of multiple market systems under different policy conditions. The power material market includes the carbon market, fossil fuel market, and new energy market. The second category is a computable general equilibrium model that uses actual economic data to predict how the economy will respond to changes in policies, technology, or other external factors. The third category is to derive price transmission by establishing a market model and analyzing market price correlation.

[0003] Current methods for calculating market price correlations focus on analyzing price correlation mechanisms in rational market models, which often lack close links with the actual operation of the market and are constrained by the limitations of these models. As a result, the ability to generalize and effectively analyze the correlation between electricity prices and power material prices in a complex market environment is relatively poor. Summary of the invention

[0004] The present invention provides a scheduling method and system for electricity, carbon and fossil fuels based on cointegration theory. The long-term equilibrium relationship and short-term dynamic adjustment characteristics between the prices of electricity, carbon and fossil fuels are analyzed based on cointegration theory, which can retain as much original information as possible, help to accurately capture the correlation between markets, optimize the scheduling of power materials according to the results of cointegration analysis, and enhance economic and environmental value.

[0005] In order to solve the above technical problems, an embodiment of the present invention provides a scheduling method for electricity, carbon and fossil fuels based on cointegration theory, including:

[0006] Obtain the market price series during the electricity trading period;

[0007] Based on the cointegration theory, the cointegration vector between the market price series is calculated;

[0008] Establishing a comprehensive power generation cost index model based on the market price sequence;

[0009] According to the cointegration vector, the comprehensive power generation cost index model is optimized to generate a scheduling plan for power materials.

[0010] As an improvement of the above solution, the cointegration vector between the market price series is calculated based on the cointegration theory, including:

[0011] Preprocessing the market price sequence to obtain the market price sequence with the same single integration order as the cointegrated market price sequence;

[0012] Calculating the optimal lag order of the cointegrated market price series according to a preset information criterion; the preset information criterion includes the Bayesian information criterion, the Hannan-Quinn criterion, the final prediction error and the Akaike information criterion;

[0013] Calculating the cointegration rank and parameter matrix of the cointegrated market price series according to the optimal lag order;

[0014] The cointegration vectors between the market price series are calculated according to the cointegration rank and the parameter matrix.

[0015] As an improvement of the above solution, the market price sequence is preprocessed to obtain the market price sequence with the same single integration order as the cointegrated market price sequence, including:

[0016] The units of the market price series are processed uniformly, and the data are supplemented by using the data smoothing method to obtain a perfect market price series;

[0017] The enhanced Dickey-fuller test is used to test the stationarity of the perfect market price series. When the test result is 0, the perfect market price series is differentiated until the perfect market price series with a test result of 1 is obtained as the cointegrated market price series. The single integration order of the cointegrated market price series is the number of differences.

[0018] As an improvement of the above solution, the cointegration rank and parameter matrix of the cointegration market price series is calculated according to the optimal lag order, including:

[0019] According to the optimal lag order, a VAR model of the cointegrated market price series is established;

[0020] The VAR model is differentiated and summed with the error correction term to establish a VECM model of the cointegrated market price series;

[0021] Solving the VECM model, obtaining the cointegration rank and parameter matrix of the cointegrated market price series.

[0022] As an improvement of the above solution, the cointegration vector between the market price series is calculated according to the cointegration rank and the parameter matrix, including:

[0023] Decomposing the parameter matrix according to the cointegration rank to obtain a cointegration vector matrix;

[0024] According to the cointegration vector matrix, the cointegration vectors between the market price sequences are obtained.

[0025] As an improvement of the above scheme, the comprehensive power generation cost index model is established according to the market price sequence, including:

[0026] Obtain the electricity trading volume, fixed cost, energy consumption function, and carbon emission function of different power materials;

[0027] A comprehensive power generation cost index model is constructed based on the market price sequence, the electricity trading volume, the fixed cost, the energy consumption function and the carbon emission function.

[0028] As an improvement of the above solution, the different power materials include coal power, oil power, gas power, and renewable energy power;

[0029] Then, constructing a comprehensive power generation cost index model according to the market price sequence, the power trading volume, the fixed cost, the energy consumption function and the carbon emission function includes:

[0030] Calculate the fuel cost based on the electricity transaction volume and energy consumption function of coal-fired power, oil-fired power and gas-fired power and the market price sequence;

[0031] Calculate fixed operating and management costs based on the electricity trading volume and fixed costs of coal-fired power, oil-fired power, gas-fired power, and renewable energy power;

[0032] Calculate the carbon cost based on the electricity trading volume of coal-fired power, oil-fired power and gas-fired power, the carbon emission function and the market price sequence;

[0033] A comprehensive power generation cost index model is constructed according to the fuel cost, the fixed operating and management cost, the carbon cost and the electricity trading volume.

[0034] As an improvement of the above scheme, the comprehensive power generation cost index model is optimized according to the cointegration vector to generate a scheduling scheme for power materials, including:

[0035] According to the cointegration vector, taking the minimum value of the comprehensive power generation cost index model as the optimization goal, the dispatching quantity of power materials is calculated;

[0036] A scheduling plan for electric power materials is generated according to the scheduling amounts of carbon and fossil fuels.

[0037] As an improvement of the above scheme, after optimizing the comprehensive power generation cost index model according to the cointegration vector to generate a scheduling scheme for power materials, the scheduling method for power materials based on the cointegration theory further includes:

[0038] Generate a correlation coefficient matrix of the market price series according to the cointegration vector;

[0039] Factoring the correlation coefficient matrix using a square root method to obtain a lower triangular matrix;

[0040] Constructing a standard normal distribution sample matrix according to the lower triangular matrix and the randomly generated standard normal distribution variables;

[0041] A simulated price sample matrix with correlation is generated according to the correlation coefficient matrix and the standard normal distribution sample matrix.

[0042] The embodiment of the present invention further provides a scheduling system for electricity, carbon and fossil fuels based on cointegration theory, including:

[0043] A market price sequence acquisition module is used to acquire the market price sequence during the power trading period;

[0044] A cointegration vector calculation module, used for calculating the cointegration vectors between the market price series based on the cointegration theory;

[0045] A model building module, used to establish a comprehensive power generation cost index model according to the market price sequence;

[0046] A scheme generation module is used to optimize the comprehensive power generation cost index model according to the cointegration vector to generate a scheduling scheme for power materials.

[0047] Compared with the prior art, the present invention discloses a method and system for dispatching electricity, carbon and fossil fuels based on cointegration theory, which obtains the market price sequence during the electricity trading period; calculates the cointegration vector between the market price sequence based on the cointegration theory; establishes a comprehensive power generation cost index model according to the market price sequence; optimizes the comprehensive power generation cost index model according to the cointegration vector to generate a dispatching plan for power materials. By adopting the embodiment of the present invention, the long-term equilibrium relationship and short-term dynamic adjustment characteristics between the prices of electricity, carbon and fossil fuels can be analyzed based on the cointegration theory, and the original information can be retained as much as possible, which helps to accurately capture the correlation between markets, optimize the dispatching of power materials according to the cointegration analysis results, and improve the economic and environmental value. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a schematic diagram of the steps of a scheduling method for electricity, carbon and fossil fuels based on cointegration theory provided by an embodiment of the present invention;

[0049] Figure 2 It is a structural schematic diagram of a scheduling system for electricity, carbon and fossil fuels based on cointegration theory provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] In the description of the specification and claims, it is to be understood that the terms first, second, etc. in the specification and claims are only used for the purpose of describing the same technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated, nor necessarily describing the order or time sequence. The terms are interchangeable where appropriate. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features.

[0052] The embodiment of the present invention provides a scheduling method for electricity, carbon and fossil fuels based on cointegration theory. Figure 1 In this embodiment, the scheduling method of electricity, carbon and fossil fuels based on cointegration theory is specifically performed through steps S1 to S4:

[0053] S1. Obtain the market price sequence during the electricity trading period.

[0054] It should be noted that the determination of electricity market prices does not depend on a single generator, but on the spatiotemporal characteristics of the comprehensive power generation costs of all power generation resources in the market coverage area. The embodiment of the present invention achieves the extraction of the spatiotemporal characteristics of the comprehensive power generation costs of power generation resources by acquiring market price sequences.

[0055] Exemplarily, the market price sequence includes an electricity price sequence and an electricity material price sequence, and the electricity material price sequence includes a carbon price sequence and a fossil fuel price sequence.

[0056] S2. Based on the cointegration theory, the cointegration vector between the market price series is calculated.

[0057] It should be noted that the market price series is usually a non-stationary time series. In order to avoid false regression when establishing a regression model for a non-stationary time series in the prior art, the non-stationary series is usually converted into a stationary time series by difference, but the difference may cause the loss of valuable information in the time series, and the final price correlation calculation result is not accurate enough. The present invention processes the market price series based on the cointegration theory, and when the non-stationary time series satisfies the same order single integration, the regression of the non-stationary time series can be effectively analyzed. Therefore, in an embodiment of the present invention, when calculating the cointegration vector, it is necessary to verify whether the market price series satisfies the same order single integration.

[0058] It should also be noted that the cointegration vector reflects the weight or ratio of the market price series in the long-term equilibrium state. The sample calculated based on the cointegration theory can retain as much original information as possible, which helps to capture the correlation between markets more accurately.

[0059] S3. Establish a comprehensive power generation cost index model based on the market price sequence.

[0060] It should be noted that the comprehensive power generation cost index model can reflect the price correlation between the prices of various types of power materials and the total cost. By adjusting the parameters in the comprehensive power generation cost index model, the power generation cost can be controlled.

[0061] S4. According to the cointegration vector, the comprehensive power generation cost index model is optimized to generate a scheduling plan for power materials.

[0062] It can be understood that the cointegration vector can reflect the long-term equilibrium relationship between the electricity and power material markets. Optimizing the comprehensive power generation cost index model based on the cointegration vector can deeply understand the dynamic interaction between markets, and the generated scheduling plan is helpful to improve economic and environmental value.

[0063] In the above scheme, the long-term equilibrium relationship and short-term dynamic adjustment characteristics between the prices of power materials are analyzed based on the cointegration theory, which can retain as much original information as possible, help to accurately capture the correlation between markets, and optimize the scheduling of power materials based on the results of cointegration analysis, which can enhance economic and environmental value.

[0064] As a preferred implementation, step S2, based on the cointegration theory, calculates the cointegration vector between the market price series, which is specifically performed through steps S21-S24:

[0065] S21. Preprocess the market price sequence to obtain the market price sequence with the same single integration order as the cointegrated market price sequence.

[0066] It should be noted that the stationarity of the market price series can be determined by checking whether the single integral orders of each market price series are the same. If the single integral orders of the market price series are not the same, the market price series needs to be further processed.

[0067] Exemplarily, as a preferred implementation, step S21 includes:

[0068] The units of the market price series are processed uniformly, and the data are supplemented by using the data smoothing method to obtain a perfect market price series;

[0069] The enhanced Dickey-fuller test is used to test the stationarity of the perfect market price series. When the test result is 0, the perfect market price series is differentiated until the perfect market price series with a test result of 1 is obtained as the cointegrated market price series. The single integration order of the cointegrated market price series is the number of differences.

[0070] It should be noted that when the test result of the Dickey-fuller test is 0, it means that the perfect market price series is not stable and it is difficult to analyze using the cointegration theory. By differencing the perfect market price series, the series can gradually tend to be stable; when the test result of the Dickey-fuller test is 1, it means that the perfect market price series is stable.

[0071] S22. Calculate the optimal lag order of the cointegrated market price series according to a preset information criterion; the preset information criterion includes the Bayesian information criterion, the Hannan-Quinn criterion, the final prediction error and the Akaike information criterion.

[0072] It should be noted that, among the preset information criteria, the Bayesian information criterion and the Hannan-Quinn criterion provide consistent estimates of the true lag order, while the final prediction error and the Akaike information criterion tend to overestimate the lag order.

[0073] In some embodiments, the preset information criterion may also include a log-likelihood function value, a likelihood ratio, or a p-value.

[0074] As an example, the maximum lag order is set to 6 and the optimal lag order is calculated, and the order containing the most minimum information criteria is selected from the calculation results as the optimal lag order p.

[0075] It should be noted that the maximum lag order is set to 6 for example only. The value of the maximum lag order is related to the complexity of the model and the amount of information it contains, and should be selected according to actual conditions. The specific value does not affect the beneficial effects achieved by the present invention.

[0076] S23. Calculate the cointegration rank and parameter matrix of the cointegrated market price series according to the optimal lag order.

[0077] For example, based on the Johansen trace test method, the cointegration rank between multiple market price series is tested. Under the r-order cointegration rank, if the test statistic is less than the statistic at the 5% significance level, it indicates that the cointegration rank between multiple market price series can be considered to be r at the 5% significance level.

[0078] Further, preferably, step S23 includes:

[0079] According to the optimal lag order, a VAR model of the cointegrated market price series is established;

[0080] The VAR model is differentiated and summed with the error correction term to establish a VECM model of the cointegrated market price series;

[0081] Solving the VECM model, obtaining the cointegration rank and parameter matrix of the cointegrated market price series.

[0082] Exemplarily, in some embodiments, the VAR model is:

[0083]

[0084] Among them, Y t is a cointegrated market price series of m yuan; A is a constant term, B i is an m×m-order parameter matrix, i=1,2,…,p; U t is the random error, U t ~N(0,σ 2 );

[0085] The VECM model is:

[0086]

[0087] Among them, Φ i and Φ are parameter matrices; Φ i For Y t Short-term adjustment for changes in Φ i =-(I-B1-…-B i );Φ is the value of Y t Long-term adjustment for changes, Φ = -(I-B1-…-B p ), the matrix Φ is of order m×m, and its rank r is the cointegration rank.

[0088] In the above scheme, the VAR model represents a long-term equilibrium model. In the real economy, elements are rarely in equilibrium, so only short-term or non-equilibrium relationships between elements are usually observed. The VECM model represents a short-term non-equilibrium model, which contains a VAR model with a difference term and an error correction term ΦY t-p The error correction term is used to introduce long-term information that is lost in the system due to the difference. Φ i It can be regarded as Y t The short-term adjustment of the change in Y t Long-term adjustments to changes.

[0089] S24. Calculate the cointegration vectors between the market price sequences according to the cointegration rank and the parameter matrix.

[0090] Preferably, step S24 includes:

[0091] Decomposing the parameter matrix according to the cointegration rank to obtain a cointegration vector matrix;

[0092] According to the cointegration vector matrix, the cointegration vectors between the market price sequences are obtained.

[0093] It should be noted that, in the embodiment of the present invention, the decomposition of the parameter matrix is ​​for Y t The matrix Φ of the long-term adjustment of the change is decomposed. Φ can be decomposed into the product of two matrices a and b, with an order of m×r; where a is the speed of non-equilibrium adjustment and b is the cointegration vector matrix; a row of the cointegration vector matrix is ​​regarded as a cointegration vector, and the cointegration rank r is the number of cointegration vectors.

[0094] It should also be noted that, in the embodiment of the present invention, a and b are not unique individually, but b uniquely defines a cointegration space, and a and b can be properly normalized.

[0095] As a preferred implementation, step S3, establishing a comprehensive power generation cost index model according to the market price sequence, includes:

[0096] Obtain the electricity trading volume, fixed cost, energy consumption function, and carbon emission function of different power materials;

[0097] A comprehensive power generation cost index model is constructed based on the market price sequence, the electricity trading volume, the fixed cost, the energy consumption function and the carbon emission function.

[0098] More preferably, the different power materials include coal power, oil power, gas power, and renewable energy power; then, constructing a comprehensive power generation cost index model according to the market price sequence, the power trading volume, the fixed cost, the energy consumption function, and the carbon emission function includes:

[0099] Calculate the fuel cost based on the electricity transaction volume and energy consumption function of coal-fired power, oil-fired power and gas-fired power and the market price sequence;

[0100] Calculate fixed operating and management costs based on the electricity trading volume and fixed costs of coal-fired power, oil-fired power, gas-fired power, and renewable energy power;

[0101] Calculate the carbon cost based on the electricity trading volume of coal-fired power, oil-fired power and gas-fired power, the carbon emission function and the market price sequence;

[0102] A comprehensive power generation cost index model is constructed according to the fuel cost, the fixed operating and management cost, the carbon cost and the electricity trading volume.

[0103] In the above scheme, the comprehensive power generation cost index model can effectively reflect the comprehensive cost of production, operation and carbon emissions of all power generation resources within the jurisdiction of a specific power market.

[0104] Exemplarily, in some embodiments, the fuel cost is calculated as:

[0105]

[0107] in, and They are the energy consumption functions of coal-fired power, oil-fired power, and gas-fired power respectively; and are the electricity trading volumes of coal-fired power, oil-fired power, and gas-fired power respectively; P coal (t), P oil (t), P gas (t) are the prices of coal-fired power, oil-fired power, and gas-fired power in period t respectively;

[0108] Calculate fixed operating and administrative costs as:

[0109]

[0110] in, They are the fixed costs of coal-fired power, oil-fired power, gas-fired power, and renewable energy power; The electricity trading volumes are coal-fired power, oil-fired power, gas-fired power, and renewable energy power respectively;

[0111] The carbon cost is calculated as:

[0112]

[0113] in, They are the carbon emission functions of coal-fired power, oil-fired power, and gas-fired power respectively; are the electricity trading volumes of coal-fired power, oil-fired power, and gas-fired power respectively; P carbon (t) is the carbon price in period t;

[0114] The constructed comprehensive power generation cost index model is:

[0115]

[0116] Among them, C fuel (t) is the fuel cost; C operate (t) is fixed operating and management cost; C carbon (t) is the carbon cost; They are respectively the electricity trading volumes of coal-fired power, oil-fired power, gas-fired power and renewable energy power.

[0117] As a preferred implementation, step S4, optimizing the comprehensive power generation cost index model according to the cointegration vector to generate a scheduling plan for power materials, includes:

[0118] According to the cointegration vector, taking the minimum value of the comprehensive power generation cost index model as the optimization goal, the dispatching quantity of power materials is calculated;

[0119] A scheduling plan for electric power materials is generated according to the scheduling amounts of carbon and fossil fuels.

[0120] In some embodiments, an optimization scheduling model is established with the minimum value of the comprehensive power generation cost index model as the optimization goal:

[0121]

[0122] Among them, C fuel (t) is the fuel cost; C operate (t) is fixed operating and management cost; C carbon (t) is the carbon cost.

[0123] In the above scheme, when solving the optimization scheduling model, the cointegration vector is taken into consideration, so that the prices of various types of power generation energy can be comprehensively considered in the process of cost optimization, and the most economical and environmentally friendly scheduling plan for power materials can be obtained.

[0124] In some embodiments, as a preferred implementation, after step S4, the power material scheduling method based on cointegration theory further includes steps S51 to S54:

[0125] S51. Generate a correlation coefficient matrix of the market price series according to the cointegration vector.

[0126] In some embodiments, the cointegration vector is α=(α1, α2, …, α k ), the correlation coefficient matrix A = diag(α i ), (i=1,2,…,k), where diag(α i ) is the diagonal matrix for extracting the cointegrating vectors.

[0127] S52. Factor decomposing the correlation coefficient matrix using a square root method to obtain a lower triangular matrix.

[0128] In some embodiments, by A=C α,t C α,t T Factor decomposition of the correlation coefficient matrix yields a lower triangular matrix

[0129] S53. Construct a standard normal distribution sample matrix according to the lower triangular matrix and the randomly generated standard normal distribution variables.

[0130] In some embodiments, by Construct a standard normal distribution sample matrix; where Z is a randomly generated standard normal distribution variable, are all N-dimensional random variables.

[0131] S54. Generate a price sample matrix with correlation according to the correlation coefficient matrix and the standard normal distribution sample matrix.

[0132] In some embodiments, by Y t =diag(σ i )Z α,t +λ mn,t Sampling the standard normal distribution sample matrix M times to obtain a price sample matrix with correlation;

[0133] Among them, σ i is the variance matrix of the standard normal distribution sample matrix; diag(σ i ) is the diagonal matrix of the variance matrix; Z α,t is the standard normal distribution sample matrix; mn,t It is an N-dimensional, N-column matrix consisting of the average values ​​of the m-th price and the n-th price in the market price sequence on the t-th day.

[0134] In the above scheme, generating a price sample matrix helps to deeply understand the dynamic interactions between markets and provides an important reference for market participants.

[0135] The method for dispatching electricity, carbon and fossil fuels based on cointegration theory provided by an embodiment of the present invention can analyze the long-term equilibrium relationship and short-term dynamic adjustment characteristics between the prices of electricity, carbon and fossil fuels based on cointegration theory, can retain as much original information as possible, and is helpful to accurately capture the correlation between markets. The method can optimize the dispatching of power materials according to the cointegration analysis results, thereby improving the economic and environmental value.

[0136] See also Figure 2 The embodiment of the present invention provides a schematic diagram of the structure of a dispatching system for electricity, carbon and fossil fuels based on cointegration theory. The dispatching system for electricity, carbon and fossil fuels based on cointegration theory includes a market price series acquisition module 11, a cointegration vector calculation module 12, a model construction module 13 and a solution generation module 14, wherein:

[0137] The market price sequence acquisition module 11 is used to acquire the market price sequence during the power trading period;

[0138] A cointegration vector calculation module 12 is used to calculate the cointegration vectors between the market price sequences based on the cointegration theory;

[0139] A model building module 13, used to establish a comprehensive power generation cost index model according to the market price sequence;

[0140] The scheme generating module 14 is used to optimize the comprehensive power generation cost index model according to the cointegration vector and generate a scheduling scheme for power materials.

[0141] As a preferred implementation, the cointegration vector calculation module 12 includes:

[0142] A preprocessing unit, used for preprocessing the market price sequence to obtain the market price sequence with the same single integration order as the cointegrated market price sequence;

[0143] An optimal lag order calculation unit, used to calculate the optimal lag order of the cointegrated market price series according to a preset information criterion; the preset information criterion includes the Bayesian information criterion, the Hannan-Quinn criterion, the final prediction error and the Akaike information criterion;

[0144] A cointegration rank and parameter matrix calculation unit, used to calculate the cointegration rank and parameter matrix of the cointegration market price series according to the optimal lag order;

[0145] The cointegration vector calculation unit is used to calculate the cointegration vector between the market price series according to the cointegration rank and the parameter matrix.

[0146] Further, preferably, the preprocessing unit is specifically used for:

[0147] The units of the market price series are processed uniformly, and the data are supplemented by using the data smoothing method to obtain a perfect market price series;

[0148] The enhanced Dickey-fuller test is used to test the stationarity of the perfect market price series. When the test result is 0, the perfect market price series is differentiated until the perfect market price series with a test result of 1 is obtained as the cointegrated market price series. The single integration order of the cointegrated market price series is the number of differences.

[0149] Preferably, the cointegration rank and parameter matrix calculation unit is specifically used for:

[0150] According to the optimal lag order, a VAR model of the cointegrated market price series is established;

[0151] The VAR model is differentiated and summed with the error correction term to establish a VECM model of the cointegrated market price series;

[0152] Solving the VECM model, obtaining the cointegration rank and parameter matrix of the cointegrated market price series.

[0153] Preferably, the cointegration vector calculation unit is specifically used for:

[0154] Decomposing the parameter matrix according to the cointegration rank to obtain a cointegration vector matrix;

[0155] According to the cointegration vector matrix, the cointegration vectors between the market price sequences are obtained.

[0156] As a preferred implementation, the model building module 13 includes:

[0157] A data acquisition unit is used to obtain the power transaction volume, fixed cost, energy consumption function, and carbon emission function of different power materials;

[0158] A model building unit is used to build a comprehensive power generation cost index model based on the market price sequence, the electricity trading volume, the fixed cost, the energy consumption function and the carbon emission function.

[0159] Further, preferably, the different power materials include coal power, oil power, gas power, and renewable energy power, then the model building unit is specifically used for:

[0160] Calculate the fuel cost based on the electricity transaction volume and energy consumption function of coal-fired power, oil-fired power and gas-fired power and the market price sequence;

[0161] Calculate fixed operating and management costs based on the electricity trading volume and fixed costs of coal-fired power, oil-fired power, gas-fired power, and renewable energy power;

[0162] Calculate the carbon cost based on the electricity trading volume of coal-fired power, oil-fired power and gas-fired power, the carbon emission function and the market price sequence;

[0163] A comprehensive power generation cost index model is constructed according to the fuel cost, the fixed operating and management cost, the carbon cost and the electricity trading volume.

[0164] As a preferred implementation, the solution generation module 14 includes:

[0165] A dispatch quantity calculation unit, used to calculate the dispatch quantity of electric power materials according to the cointegration vector and taking the minimum value of the comprehensive power generation cost index model as the optimization target;

[0166] A plan generating unit is used to generate a scheduling plan for electric power materials according to the scheduling amount of carbon and fossil fuels.

[0167] As a preferred implementation, the power material dispatching system based on cointegration theory further includes:

[0168] A correlation coefficient matrix generation module, used for generating a correlation coefficient matrix of the market price sequence according to the cointegration vector;

[0169] A factor decomposition module, used for factorizing the correlation coefficient matrix by using a square root method to obtain a lower triangular matrix;

[0170] A standard normal distribution sample matrix construction module is used to construct a standard normal distribution sample matrix according to the lower triangular matrix and randomly generated standard normal distribution variables;

[0171] The simulated price sample matrix generation module is used to generate a simulated price sample matrix with correlation according to the correlation coefficient matrix and the standard normal distribution sample matrix.

[0172] The scheduling system for electricity, carbon and fossil fuels based on cointegration theory provided by the embodiment of the present invention can analyze the long-term equilibrium relationship and short-term dynamic adjustment characteristics between the prices of electricity, carbon and fossil fuels based on cointegration theory, can retain as much original information as possible, and help to accurately capture the correlation between markets. It can optimize the scheduling of power materials according to the results of cointegration analysis, and can enhance economic and environmental value.

[0173] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0174] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A scheduling method for electricity, carbon and fossil fuels based on cointegration theory, characterized in that: include: Obtain the market price series during the electricity trading period; Based on the cointegration theory, the cointegration vector between the market price series is calculated; Establishing a comprehensive power generation cost index model based on the market price sequence; According to the cointegration vector, the comprehensive power generation cost index model is optimized to generate a scheduling plan for power materials.

2. A method for dispatching electricity, carbon and fossil fuels based on cointegration theory as claimed in claim 1, characterized in that: The cointegration vector between the market price series is calculated based on the cointegration theory, including: Preprocessing the market price sequence to obtain the market price sequence with the same single integration order as the cointegrated market price sequence; Calculating the optimal lag order of the cointegrated market price series according to a preset information criterion; the preset information criterion includes the Bayesian information criterion, the Hannan-Quinn criterion, the final prediction error and the Akaike information criterion; Calculating the cointegration rank and parameter matrix of the cointegrated market price series according to the optimal lag order; The cointegration vectors between the market price series are calculated according to the cointegration rank and the parameter matrix.

3. A method for dispatching electricity, carbon and fossil fuels based on cointegration theory as claimed in claim 2, characterized in that: The preprocessing of the market price sequence to obtain the market price sequence with the same single integration order as the cointegrated market price sequence includes: The units of the market price series are processed uniformly, and the data are supplemented by using the data smoothing method to obtain a perfect market price series; The enhanced Dickey-fuller test is used to test the stationarity of the perfect market price series. When the test result is 0, the perfect market price series is differentiated until the perfect market price series with a test result of 1 is obtained as the cointegrated market price series. The single integration order of the cointegrated market price series is the number of differences.

4. A method for dispatching electricity, carbon and fossil fuels based on cointegration theory as claimed in claim 2, characterized in that: The step of calculating the cointegration rank and parameter matrix of the cointegrated market price series according to the optimal lag order includes: According to the optimal lag order, a VAR model of the cointegrated market price series is established; The VAR model is differentiated and summed with the error correction term to establish a VECM model of the cointegrated market price series; Solving the VECM model, obtaining the cointegration rank and parameter matrix of the cointegrated market price series.

5. The method for dispatching electricity, carbon and fossil fuels based on cointegration theory according to claim 2, characterized in that: The step of calculating the cointegration vectors between the market price sequences according to the cointegration rank and the parameter matrix comprises: Decomposing the parameter matrix according to the cointegration rank to obtain a cointegration vector matrix; According to the cointegration vector matrix, the cointegration vectors between the market price sequences are obtained.

6. The method for dispatching electricity, carbon and fossil fuels based on cointegration theory according to claim 1, characterized in that: The step of establishing a comprehensive power generation cost index model according to the market price sequence comprises: Obtain the electricity trading volume, fixed cost, energy consumption function, and carbon emission function of different power materials; A comprehensive power generation cost index model is constructed based on the market price sequence, the electricity trading volume, the fixed cost, the energy consumption function and the carbon emission function.

7. A method for dispatching electricity, carbon and fossil fuels based on cointegration theory as claimed in claim 6, characterized in that: The different power materials include coal power, oil power, gas power, and renewable energy power; Then, constructing a comprehensive power generation cost index model according to the market price sequence, the power trading volume, the fixed cost, the energy consumption function and the carbon emission function includes: Calculate the fuel cost based on the electricity transaction volume and energy consumption function of coal-fired power, oil-fired power and gas-fired power and the market price sequence; Calculate fixed operating and management costs based on the electricity trading volume and fixed costs of coal-fired power, oil-fired power, gas-fired power, and renewable energy power; Calculate the carbon cost based on the electricity trading volume of coal-fired power, oil-fired power and gas-fired power, the carbon emission function and the market price sequence; A comprehensive power generation cost index model is constructed according to the fuel cost, the fixed operating and management cost, the carbon cost and the electricity trading volume.

8. The method for dispatching electricity, carbon and fossil fuels based on cointegration theory according to claim 1, characterized in that: The step of optimizing the comprehensive power generation cost index model according to the cointegration vector to generate a scheduling plan for power materials includes: According to the cointegration vector, taking the minimum value of the comprehensive power generation cost index model as the optimization goal, the dispatching quantity of power materials is calculated; A scheduling plan for electric power materials is generated according to the scheduling amounts of carbon and fossil fuels.

9. The method for dispatching electricity, carbon and fossil fuels based on cointegration theory according to claim 1, characterized in that: After optimizing the comprehensive power generation cost index model according to the cointegration vector to generate a scheduling plan for power materials, the scheduling method for power materials based on cointegration theory further includes: Generate a correlation coefficient matrix of the market price series according to the cointegration vector; Factoring the correlation coefficient matrix using a square root method to obtain a lower triangular matrix; Constructing a standard normal distribution sample matrix according to the lower triangular matrix and the randomly generated standard normal distribution variables; A simulated price sample matrix with correlation is generated according to the correlation coefficient matrix and the standard normal distribution sample matrix.

10. A dispatching system for electricity, carbon and fossil fuels based on cointegration theory, characterized in that: include: A market price sequence acquisition module is used to acquire the market price sequence during the power trading period; A cointegration vector calculation module, used for calculating the cointegration vectors between the market price series based on the cointegration theory; A model building module, used to establish a comprehensive power generation cost index model according to the market price sequence; A scheme generation module is used to optimize the comprehensive power generation cost index model according to the cointegration vector to generate a scheduling scheme for power materials.

Images