Electricity selling company bidding decision optimization method and system based on multi-source data fusion

Abstract

The invention belongs to the technical field of electricity market transaction, and discloses an electricity selling company bidding decision optimization method and system based on multi-source data fusion, and the method comprises the steps: collecting historical electricity price data, electricity consumption data and power generation data, and carrying out the preprocessing of the collected data; constructing an electricity market environment model based on the preprocessed data; constructing a target function by taking profit maximization of the electricity selling company as a target, and constructing constraint conditions meeting the model; and solving the model based on linear programming, stochastic programming and robust optimization algorithms to obtain an optimal power purchase strategy. According to the invention, the comprehensive and refined electricity market environment model is constructed, multiple factors are comprehensively considered, and advanced data analysis technology and optimization algorithm are introduced, so that the market change trend can be accurately predicted, bidding strategies adapted to different market situations can be formulated, the bidding decision level of electricity selling companies can be effectively improved, and the market competitiveness can be improved. And the anti-risk capability and the comprehensive competitiveness in the electricity market are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of electricity market trading, and particularly to an optimization method and system for the bidding decision-making of electricity selling companies based on multi-source data fusion. Background Art

[0002] The statements in this part merely provide background technical information related to the present invention and do not necessarily constitute prior art.

[0003] With the in-depth adjustment of the global energy structure and the continuous promotion of the electricity market reform, the electricity selling market is facing an unprecedentedly complex pattern.

[0004] The proportion of renewable energy (such as wind energy, solar energy, etc.) in power supply is increasing day by day. However, its inherent volatility, intermittency, and uncertainty have brought huge challenges to the operation of electricity selling companies. When formulating bidding strategies, electricity selling companies are difficult to accurately estimate the available new energy electricity, which affects the scientificity and rationality of their power procurement plans. At the same time, the electricity consumption behaviors of different user groups (such as industrial, commercial, and residential users) are significantly different, with different electricity consumption times, electricity consumption amounts, and electricity consumption patterns, further increasing the difficulty for electricity selling companies to predict market demand. In such a complex market environment, the bidding decision-making of electricity selling companies has become extremely difficult. On the one hand, when making power purchase decisions in the medium- and long-term markets and spot markets, electricity selling companies need to comprehensively consider various factors, including the price trends of different energy sources, supply stability, policy orientation, and changes in user demands, etc. However, due to the incompleteness and uncertainty of market information, electricity selling companies are difficult to accurately grasp the dynamic relationships among various factors, resulting in huge risks in market transactions. On the other hand, most of the existing decision support systems for electricity selling companies lack comprehensive adaptability and dynamic response capabilities to complex market environments. Many systems fail to fully consider the uncertainty of new energy and user demands during model construction, and it is also difficult to track market changes in real time and make timely adjustments in algorithm optimization. This makes electricity selling companies unable to quickly optimize their bidding strategies when facing market fluctuations, missing market opportunities or increasing operating costs. In addition, the continuous changes in the policy environment, such as the adjustment of renewable energy subsidy policies and the introduction of carbon emission trading systems, etc., have also brought more uncertainties to the operation of electricity selling companies. To sum up, against the background of the increasing proportion of new energy, complex and variable electricity consumption demands, and continuous adjustment of the policy environment, there is an urgent need for an innovative optimization method for the bidding decision-making of electricity selling companies to improve the market adaptability, operating efficiency, and profitability of electricity selling companies, and ensure the stable and efficient operation of the electricity market. Summary of the Invention

[0005] To solve the above problems, the present invention proposes an optimization method and system for the bidding decision-making of electricity selling companies based on multi-source data fusion. By constructing a comprehensive and refined electricity market environment model, considering various factors, and introducing advanced data analysis techniques and optimization algorithms, it can accurately predict market change trends, formulate bidding strategies suitable for different market situations, effectively improve the bidding decision-making level of electricity selling companies, enhance their risk resistance and comprehensive competitiveness in the electricity market, and promote the efficient and stable operation of the electricity market.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] In the first aspect, the present invention provides an optimization method for the bidding decision-making of electricity selling companies based on multi-source data fusion, including the following steps:

[0008] Collect historical electricity price data, electricity consumption data, and power generation data, and preprocess the collected data;

[0009] Construct an electricity market environment model based on the preprocessed data;

[0010] Construct an objective function with the maximization of the profit of the electricity selling company as the goal and construct constraint conditions that satisfy the model;

[0011] Solve the model based on linear programming, stochastic programming, and robust optimization algorithms to obtain the optimal power purchase strategy.

[0012] As an alternative implementation, preprocessing the collected data includes data cleaning, data normalization, and data filling.

[0013] As an alternative implementation, the electricity market environment model includes an electricity price prediction model, an electricity consumption demand model, and a power generation supply model.

[0014] As an alternative implementation, a method combining the ARIMA model and neural network is used to construct the electricity price prediction model; a multiple linear regression method is used to construct the electricity consumption demand model; and a power generation supply model is constructed considering cost and policy impacts.

[0015] As an alternative implementation, the objective function is:

[0016]

[0017] Wherein, P sell,t is the electricity selling price at each time period, L t is the electricity consumption of users, C e,t is the power purchase cost, C g,t is the green certificate cost, C risk,t is the risk cost.

[0018] As an alternative implementation, the constraint conditions include power balance constraint, contract power constraint, market rule constraint, and renewable energy quota constraint.

[0019] In a second aspect, the present invention provides an optimization system for the bidding decision of an electricity selling company based on multi-source data fusion, including:

[0020] A data collection and preprocessing module, configured to: collect historical electricity price data, electricity consumption data, and power generation data, and preprocess the collected data;

[0021] A model construction module, configured to: construct a power market environment model based on the preprocessed data;

[0022] A target function and constraint condition construction module, configured to: construct a target function with the maximization of the profit of the electricity selling company as the target and construct constraint conditions that satisfy the model;

[0023] A model solving module, configured to: solve the model based on linear programming, stochastic programming, and robust optimization algorithms to obtain an optimal power purchase strategy.

[0024] In a third aspect, the present invention provides an electronic device, including a memory, a processor, and computer instructions stored on the memory and running on the processor. When the computer instructions are run by the processor, the method described in the first aspect is completed.

[0025] In a fourth aspect, the present invention provides a computer-readable storage medium for storing computer instructions. When the computer instructions are executed by a processor, the method described in the first aspect is completed.

[0026] In a fifth aspect, the present invention provides a computer program product, including a computer program. When the computer program is executed by a processor, the method described in the first aspect is implemented.

[0027] Compared with the prior art, the beneficial effects of the present invention are:

[0028] The present invention provides an optimization method and system for the bidding decision of electricity selling companies based on multi-source data fusion, which solves the problem that traditional bidding decision methods often ignore the dynamic changes and mutual influences of many key factors in the market, making it difficult for electricity selling companies to formulate optimal strategies in a complex environment. By constructing a comprehensive and refined electricity market environment model, comprehensively considering various factors such as the characteristics of new energy power generation, the diversity of user demands, the dynamics of policies and regulations, and the market competition situation, and at the same time introducing advanced data analysis technologies and optimization algorithms, the present invention can accurately predict the market change trend, formulate bidding strategies suitable for different market situations, effectively improve the bidding decision-making level of electricity selling companies, enhance their anti-risk ability and comprehensive competitiveness in the electricity market, and promote the efficient and stable operation of the electricity market. It has the following advantages:

[0029] (1) In terms of data processing and decision-making foundation construction, it innovatively integrates multi-source data fusion technology: by widely collecting data from multiple channels such as electricity market trading platforms, power generation enterprises, user electricity meters, and meteorological monitoring, and using fusion algorithms such as weighted average, it deeply integrates data from different sources and types, enabling electricity selling companies to obtain more comprehensive and accurate market information. Based on the decision-making model constructed from the fused data, electricity selling companies can make decisions that are more in line with the actual market situation when formulating bidding strategies.

[0030] (2) The characteristics of strategy adaptability and stability guaranteed by the hybrid optimization algorithm: aiming at the uncertain factors of the complex and changeable market, it uses a hybrid optimization algorithm, combines advanced algorithms such as stochastic programming and robust optimization, and constructs a decision optimization framework suitable for different market situations. The hybrid optimization algorithm of the present invention can make the bidding strategies of electricity selling companies have stronger adaptability and robustness, and ensure their operational stability.

[0031] (3) The improvement of efficient decision-making and market response ability realized by the cooperation of the model and the algorithm: in terms of the cooperation between model construction and the algorithm, the constructed electricity market environment model (including electricity price prediction, electricity demand, and power generation supply models, etc.) is closely combined with the hybrid optimization algorithm. This synergistic effect enables electricity selling companies to quickly respond to market changes in a short time and adjust their bidding strategies in a timely manner. Compared with traditional decision-making methods, the model-algorithm cooperation system of the present invention greatly improves the decision-making efficiency and market response speed, enabling electricity selling companies to achieve sustainable development and providing strong support for the stable and efficient operation of the electricity market.

[0032] The advantages of the additional aspects of the present invention will be partially given in the following description, partially will become obvious from the following description, or will be understood through the practice of the present invention. Description of the Drawings

[0033] The accompanying drawings forming a part of this specification are used to provide a further understanding of the present invention. The schematic embodiments and descriptions thereof of the present invention are used to explain the present invention and do not constitute an improper limitation of the present invention.

[0034] Figure 1 It is a flowchart of an optimization method for the bidding decision of an electricity selling company based on multi-source data fusion provided in Embodiment 1 of the present invention. Detailed implementation manners

[0035] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0036] It should be noted that the following detailed descriptions are all exemplary and are intended to provide further explanations of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0037] It should be noted that the terms used herein are only for describing specific implementation manners and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless otherwise clearly specified in the context, the singular form is also intended to include the plural form. In addition, it should also be understood that the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those clearly listed steps or units, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0038] In the case of no conflict, the embodiments in the present invention and the features in the embodiments can be combined with each other.

[0039] Embodiment 1

[0040] As Figure 1 shown, this embodiment provides an optimization method for the bidding decision of an electricity selling company based on multi-source data fusion, including the following steps:

[0041] S1. Collect historical electricity price data, electricity consumption data and power generation data, and preprocess the collected data.

[0042] S2. Construct a power market environment model based on the preprocessed data.

[0043] S3. Construct an objective function with the maximization of the profit of the electricity selling company as the goal and construct constraint conditions that satisfy the model.

[0044] S4. Solve the model based on linear programming, stochastic programming and robust optimization algorithms to obtain an optimal power purchase strategy.

[0045] First, S1: Collect historical electricity price data, electricity consumption data, and power generation data, and preprocess the collected data.

[0046] The present invention establishes a multi-source data acquisition model based on data normalization and linear interpolation method. The construction of this model aims to achieve two main objectives: one is to collect historical and latest electricity price data (including medium- and long-term markets and spot markets) in real time through a dedicated interface with the power market trading platform, and perform preliminary format conversion and collation on the data to make it meet the internal data processing requirements of the system; the other is to interact with the data systems of power generation enterprises to obtain data such as conventional energy power generation, renewable energy power generation, and the operating status of energy storage devices, so as to comprehensively grasp the supply capacity and dynamic changes on the power generation side. The specific data screening and classification steps are as follows:

[0047] Step 1: Data acquisition. Establish data interfaces with external data sources such as the power market trading platform, power generation enterprise data systems, and user electricity meter data acquisition systems to ensure timely and accurate data acquisition.

[0048] (1) Data acquisition. Let the medium- and long-term market electricity price data sequence collected from the power market trading platform be where T is the total duration of the historical data collected, and the acquisition frequency is Δt (for example, if data is collected every hour, then Δt = 1 hour). Then the medium- and long-term market electricity price P at time t L,t can be expressed as:

[0049] P L,t = p L,t (collected from the trading platform)(1)

[0050] Similarly, for the spot market electricity price data sequence the spot market electricity price P at time t S,t is:

[0051] P S,t = p S,t (collected from the trading platform)(2)

[0052] (2) Regarding user electricity consumption data, let the electricity consumption of the i-th type of user (such as residential users, industrial and commercial users, etc.) at time t be L i,t , then the total user load L t can be obtained by weighted summation of the electricity consumption of various types of users (assuming there are n types of users, with weights ω i , satisfying ):

[0053]

[0054] (3) The conventional energy power generation data sequence obtained from power generation enterprises and the data series of renewable energy power generation The conventional energy power generation G at time t f,t and the renewable energy power generation G r,t are directly collected as follows:

[0055] G f,t = g f,t (Data acquisition system for power producers) (4)

[0056] G r,t = g r,t (Data acquisition system for power producers) (5)

[0057] (4) For the policy parameters related to the new power reform policies, such as the renewable energy quota ratio θ, their values are directly obtained by analyzing the policy documents.

[0058] Step 2: Data preprocessing. Preprocess the collected data, including data cleaning (removing outliers and incorrect data), data normalization (making different types of data comparable), and data filling (using appropriate interpolation methods to complete missing data) to ensure the accuracy and availability of the data and provide a solid data foundation for subsequent model construction and analysis.

[0059] (1) Data cleaning (removing outliers): Let the mean of P L,t be μp L , and the standard deviation be σp L . If P L,t satisfies:

[0060] |P L,t - μp L |> kσp L (6)

[0061] where k is a threshold set according to experience, such as 3. Then P L,t is considered an outlier and is removed. The updated data series of the medium- and long-term market electricity price is (T' is the data duration after cleaning). For other data (such as P S,t , L t , G f,t , G r,t , etc.), similar methods are used for outlier processing.

[0062] (2) Data normalization (min-max normalization): For the medium- and long-term market electricity price data, the normalized (denoted as ) calculation formula is:

[0063]

[0064] Similarly, data such as spot market electricity prices, user loads, conventional energy power generation, and renewable energy power generation are normalized to obtain a normalized sequence.

[0065] (3) Data filling (using linear interpolation method): If a certain data (such as ) is found to be missing at time t1, let the previous known data time adjacent to it be t0, and the next known data time be t2, then the filling value of the missing data is:

[0066]

[0067] The missing values of other types of data are filled using the same linear interpolation method to ensure data integrity.

[0068] S2. Construct a power market environment model based on the preprocessed data.

[0069] Based on the preprocessed historical data, use advanced data analysis techniques and algorithms to construct a power market environment model. The power market environment model includes an electricity price prediction model, an electricity demand model, and a power generation supply model.

[0070] Step 1: Electricity price prediction model. Use a method combining the ARIMA model and neural network to construct an electricity price prediction model.

[0071] (1) First, perform ARIMA model modeling on medium- and long-term market electricity price data to determine the order of the ARIMA model (set as p, d, q). Determine the values of p and q through autocorrelation function (ACF) and partial autocorrelation function (PACF) analysis. d is the number of differencing times to make the processed time series stationary. The expression of the ARIMA model is:

[0072]

[0073] Among them, is the autoregressive polynomial, θ(B) = 1 + θ1B + θ2B 2 +…+ θ q B q is the moving average polynomial, B is the lag operator, and ε t is a white noise sequence.

[0074] (2) Then, take the prediction result of the ARIMA model as one of the input features, and construct a neural network model in combination with other influencing factors (such as user load, conventional energy power generation, renewable energy power generation, etc.). Assume that the neural network has m input layer nodes (including the electricity price predicted by ARIMA and the normalized values of other influencing factors), h hidden layer nodes (using a suitable activation function, such as the ReLU function), and 1 output layer node (the predicted medium- and long-term market electricity price at the next moment). The weight matrices of the neural network are W1 (from the input layer to the hidden layer) and W2 (from the hidden layer to the output layer), and the bias vectors are b1 (hidden layer) and b2 (output layer). Then, the medium- and long-term market electricity price prediction model is:

[0075]

[0076] Among them, f2 and f3 are the activation functions of the hidden layer and the output layer respectively, is the electricity price at time t predicted by the ARIMA model. For the spot market electricity price prediction model, it is constructed in a similar way, except that the input features and model parameters may be different.

[0077] Step 2: User electricity demand model. Use the multiple linear regression method to construct the user electricity demand model.

[0078] Assume that the user electricity demand is related to multiple factors, such as seasonal factors (denoted by s t , which can be encoded through months or quarters, etc.), time factors (such as the time period h t of a day), economic development indicators (such as the regional GDP growth rate e t ), and historical electricity consumption (such as the electricity consumption L t-1 at the previous moment, etc.). Then, the user electricity demand model can be expressed as:

[0079] L t =β0 + β1s t +β2h t +β3e t +β4L t-1 +…+ε t (11)

[0080] Among them, β0, β1, β2, … are regression coefficients, which are estimated according to historical data through the least squares method or other regression algorithms, and ε t is the error term.

[0081] Step 3: Power generation supply model, which is constructed considering the influence of cost and policy.

[0082] (1) The cost function C f,t of conventional energy power generators can be expressed as:

[0083]

[0084] Among them, a0, a1, and a2 are cost coefficients, and ΔC p,t is the cost change brought about by policy adjustment (for example, according to the policy on the market-oriented reform of the on-grid electricity price of coal-fired power generation in the new electricity reform policy, calculate the cost change caused by the electricity price floating).

[0085] (2) The cost function C of renewable energy power generators r,t Consider the power generation cost and energy storage cost (assuming the charge and discharge power of energy storage is P es,t ):

[0086] C r,t = k r G r,t + k es P es,t (13)

[0087] Among them, k r is the operation and maintenance cost coefficient of renewable energy power generation, and k es is the operation and maintenance cost coefficient of energy storage.

[0088] (3) The supply strategy of power generators is adjusted according to cost and market price. When the market electricity price is higher than the cost of power generators, power generators have the incentive to increase power generation; otherwise, they may reduce power generation. The supply model can be expressed as:

[0089] G f,t+1 = g1(P L,t+1 , P S,t+1 , C f,t , …) (14)

[0090] G r,t+1 = g2(P L,t+1 , P S,t+1 , C r,t , …) (15)

[0091] Among them, g1 and g2 are functions constructed based on the economic decision-making behavior of power generators, and the specific form can be obtained through the analysis of the historical data of power generators and economic theory modeling.

[0092] S3. Construct an objective function with the goal of maximizing the profit of the electricity sales company and construct the constraint conditions that satisfy the model.

[0093] Comprehensively integrate key elements such as the comprehensive power purchase cost, electricity sales revenue, green certificate cost, and risk cost, construct a comprehensive, accurate, and effective objective function that can guide the electricity sales company to pursue profit maximization, and set constraint conditions to ensure the rationality and feasibility of the decision-making.

[0094] Step 1: Determine the objective function of the electricity sales company.

[0095] With the goal of maximizing the profit of the electricity sales company, a comprehensive and detailed profit function is constructed. The profit function I of the electricity sales company s Comprehensively considers multiple key factors such as power purchase cost, electricity sales revenue, green certificate cost, and risk cost. The specific expression is as follows:

[0096]

[0097] Among them, the electricity sales revenue part is the sum of the products of the electricity sales price P sell,t and the electricity consumption L of users t . The power purchase cost includes the power purchase expenditures in the medium- and long-term market and the spot market, and its calculation method is:

[0098] C e,t =p′ s1,t E′ s1,t +p′ s2,t E′ s2,t +p″ s,t E″ s,t (17)

[0099] Among them, P′ s1,t ,P′ s2,t respectively represent the medium- and long-term contract electricity prices for the electricity sales company to bundle renewable energy (RES) power and conventional energy (CES) power; E′ s1,t ,E′ s2,t are the corresponding medium- and long-term contract electricity quantities; p″ s,t ,E″ s,t are the power purchase price and power purchase quantity of the electricity sales company in the spot market. The green certificate cost is determined according to the renewable energy quota system and the green certificate trading mechanism. The calculation formula is as follows:

[0100]

[0101] Among them, p g,t is the wholesale price of the electricity sales company in the green certificate market. The risk cost is calculated by evaluating market uncertainty factors (such as electricity price fluctuations, user demand changes, etc.). For example, the conditional value at risk (CVaR) method is used. The specific formula is:

[0102] C risk,t =αCVaR t (19)

[0103] Among them, α is the risk aversion coefficient, and the calculation of CVaR t is based on the risk loss distribution at a certain confidence level and is obtained through statistical analysis and simulation calculation of historical data.

[0104] Step 2: Analyze the constraint conditions. The constraint conditions include power balance constraint, contract power constraint, market rule constraint, and renewable energy quota constraint.

[0105] (1) The power balance constraint is a basic requirement to ensure the stable operation of the power market. The power purchase quantity of the electricity selling company at each time period must be equal to the power selling quantity to meet user demands and maintain the power supply-demand balance. Its mathematical expression is:

[0106] E' s1,t +E′ s2,t +E″ s,t =L t (20)

[0107] (2) The contract power constraint ensures the feasibility and enforceability of the medium- and long-term contracts signed between the electricity selling company and the power generation company. The upper and lower limits of the power quantity specified in the contract need to be considered. The specific constraint conditions are as follows:

[0108]

[0109] Among them, E′ s1,min ,E′ s1,max are the lower and upper limits of the medium- and long-term contract power quantity of RES electricity; E′ s ' 2,min ,E′ s2,max are the lower and upper limits of the medium- and long-term contract power quantity of CES electricity.

[0110] (3) The market rule constraint requires the electricity selling company to follow the trading rules of the power market, including the upper and lower limits of the quotation and trading time limits, etc. It is specifically expressed as:

[0111]

[0112] Among them, p′ s1,min ,p′ s1,max are the lower and upper limits of the medium- and long-term contract electricity price of RES electricity; p′ s2,min ,p′ s2,max are the lower and upper limits of the medium- and long-term contract electricity price of CES electricity; p″ s,min ,p″ s,max are the lower and upper limits of the electricity purchase price in the spot market.

[0113] (4) The renewable energy quota constraint is based on policy requirements to ensure that the electricity selling company meets a certain proportion of renewable energy power supply. It can be achieved by purchasing green certificates or directly purchasing renewable energy power. Its constraint condition is:

[0114]

[0115] S4. Solve the model based on linear programming, stochastic programming, and robust optimization algorithms to obtain the optimal power purchase strategy.

[0116] Step 1: Solve the optimal power purchase strategy. According to the constructed power market environment model, objective function, and constraint conditions, use an optimization algorithm to solve the optimal power purchase strategy of the electricity selling company in the medium- and long-term market and the spot market. If linear programming is used for solving, construct the following linear programming model:

[0117]

[0118] s.t. Constraint conditions (20)-(23) and the non-negativity constraints of relevant electricity prices and electricity quantities. During the solving process, make full use of the high efficiency and stability of the linear programming algorithm, and use professional optimization software packages (such as CPLEX, GUROBI, etc.) for calculation to ensure obtaining the optimal solution quickly and accurately.

[0119] Step 2: Consider the coordination of different markets. During the solving process of the optimization algorithm, fully consider the coordination effect between different markets. By analyzing the predicted results of the electricity price trends in the medium- and long-term market and the spot market and the user demand forecast situation, reasonably allocate the power purchase ratio in the two markets. For example, when it is predicted that the electricity price in the medium- and long-term market is relatively stable and lower than the expected average electricity price in the spot market, appropriately increase the medium- and long-term contract electricity quantity; conversely, if there is a low-price opportunity in the spot market in the short term and the electricity quantity balance and other constraint conditions are met, the power purchase quantity in the spot market can be increased.

[0120] Step 3: Handle uncertain factors. For the uncertain factors in the market, adopt advanced methods such as stochastic programming or robust optimization to handle them to improve the robustness and adaptability of the bidding strategy.

[0121] (1) If stochastic programming is used, assume that the uncertain parameters (such as electricity price, user load, etc.) follow a certain probability distribution (such as normal distribution, uniform distribution, etc.). Taking the electricity price as an example, let P S " ,t follow the normal distribution N(μp" S,t ,σ 2 p" S,t ), then consider the expected value of the objective function under multiple scenarios in stochastic programming:

[0122]

[0123] where S is the number of scenarios, π s is the probability of the s-th scenario, is the profit of the electricity selling company in the s-th scenario. The profit calculation for each scenario needs to substitute the actual electricity price and other parameters in this scenario into the objective function (16) for calculation. By generating a large number of random scenarios (such as thousands or even tens of thousands of scenarios) to simulate the uncertainty of the market, solve the optimization problem in each scenario, and then obtain the expected profit by weighted averaging according to the scenario probability, so as to determine the optimal decision-making strategy.

[0124] (2) If robust optimization is adopted, it is to find a solution with good performance for all possible uncertainty situations within a certain uncertainty set. For example, for the uncertainty of user load, let the uncertainty set of L t be Ω Lt , then the robust optimization model can be expressed as:

[0125]

[0126] s.t. Constraint conditions (20)-(23) and for any all satisfy certain constraint conditions (such as the robust form of the power balance constraint under uncertainty). By constructing a robust optimization model, the uncertainty factors are transformed into constraint conditions to ensure that the decision can meet certain performance requirements even in the worst case, effectively reducing risks.

[0127] Embodiment 2

[0128] This embodiment provides an optimization system for the bidding decision of an electricity selling company based on multi-source data fusion, including:

[0129] A data collection and preprocessing module, configured to: collect historical electricity price data, electricity consumption data, and power generation data, and preprocess the collected data;

[0130] A model construction module, configured to: construct a power market environment model based on the preprocessed data;

[0131] An objective function and constraint condition construction module, configured to: construct an objective function with the maximization of the profit of the electricity selling company as the goal and construct constraint conditions that satisfy the model;

[0132] A model solving module, configured to: solve the model based on linear programming, stochastic programming, and robust optimization algorithms to obtain the optimal power purchase strategy.

[0133] It should be noted here that the above modules correspond to the steps described in Embodiment 1, and the examples and application scenarios implemented by the above modules and the corresponding steps are the same, but are not limited to the content disclosed in the above Embodiment 1. It should be noted that the above modules, as part of the system, can be executed in a computer system such as a set of computer executable instructions.

[0134] In more embodiments, the following is also provided:

[0135] An electronic device, comprising a memory and a processor, and computer instructions stored on the memory and running on the processor. When the computer instructions are run by the processor, the method described in Embodiment 1 is completed. For the sake of brevity, it will not be elaborated here.

[0136] It should be understood that in this embodiment, the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), off-the-shelf programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0137] The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A part of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the device type.

[0138] A computer-readable storage medium for storing computer instructions. When the computer instructions are executed by the processor, the method described in Embodiment 1 is completed.

[0139] The method in Embodiment 1 can be directly embodied as being executed and completed by a hardware processor, or by a combination of hardware and software modules in the processor. The software module may be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. This storage medium is located in the memory, and the processor reads the information in the memory and combines its hardware to complete the steps of the above method. To avoid repetition, it will not be described in detail here.

[0140] A computer program product, comprising a computer program. When the computer program is executed by the processor, the method described in Embodiment 1 is implemented and completed.

[0141] The present invention also provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer-executable instructions, such as instructions included in program modules, which are executed in a device on a target real or virtual processor to execute the process / method as described above. Generally, program modules include routines, programs, libraries, objects, classes, components, data structures, etc. that perform specific tasks or implement specific abstract data types. In various embodiments, the functions of program modules can be combined or divided as needed. The machine-executable instructions for program modules can be executed locally or within a distributed device. In a distributed device, program modules can be located in local and remote storage media.

[0142] The computer program code for implementing the method of the present invention can be written in one or more programming languages. These computer program codes can be provided to the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing devices, such that when the program code is executed by the computer or other programmable data processing devices, the functions / operations specified in the flowchart and / or block diagram are implemented. The program code can be executed entirely on the computer, partially on the computer, as an independent software package, partially on the computer and partially on a remote computer, or entirely on a remote computer or server.

[0143] In the context of the present invention, the computer program code or related data can be carried by any suitable carrier so that the device, apparatus, or processor can perform the various processes and operations described above. Examples of the carrier include signals, computer-readable media, and so on. Examples of signals can include electrical, optical, radio, acoustic, or other forms of propagated signals, such as carrier waves, infrared signals, etc.

[0144] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in conjunction with this embodiment can be implemented by electronic hardware or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0145] Although the specific implementation manners of the present invention have been described above in conjunction with the accompanying drawings, it is not a limitation to the protection scope of the present invention. Those skilled in the art should understand that based on the technical solution of the present invention, various modifications or deformations that can be made by those skilled in the art without creative efforts are still within the protection scope of the present invention.

Claims

1. A bidding decision optimization method for power sales companies based on multi-source data fusion, characterized in that: The following steps are involved: Collect historical electricity price data, electricity consumption data and power generation data, and pre-process the collected data; Constructing the power market environment model based on the preprocessed data; Constructing an objective function with the goal of maximizing the profit of the power sales company and constructing constraints that satisfy the model; The model is solved based on linear programming, stochastic programming and robust optimization algorithms to obtain the optimal electricity purchasing strategy.

2. The bidding decision optimization method for power sales companies based on multi-source data fusion according to claim 1 is characterized in that: Preprocess the collected data, including data cleaning, data normalization and data filling.

3. The bidding decision optimization method for power sales companies based on multi-source data fusion according to claim 1 is characterized in that: The power market environment model includes an electricity price prediction model, a power demand model and a power generation supply model.

4. The bidding decision optimization method for power sales companies based on multi-source data fusion according to claim 1 is characterized in that: The ARIMA model is combined with the neural network to build an electricity price forecasting model. The multivariate linear regression method is used to build an electricity demand model. The power generation supply model is built by considering the cost and policy impact.

5. The bidding decision optimization method for power sales companies based on multi-source data fusion according to claim 1 is characterized in that: The objective function is: Among them, P sell,t is the electricity price in each period, L t is the user's electricity consumption, C e,t is the electricity purchase cost, C g,t is the cost of green certificate, C risk,t For risk cost.

6. The method for optimizing bidding decisions of power sales companies based on multi-source data fusion according to claim 1, characterized in that: Constraints include electricity balance constraints, contractual electricity constraints, market rule constraints and renewable energy quota constraints.

7. The bidding decision optimization system for power sales companies based on multi-source data fusion is characterized by: include: The data collection and preprocessing module is configured to: collect historical electricity price data, electricity consumption data and power generation data, and preprocess the collected data; The model building module is configured to: build a power market environment model based on the preprocessed data; The objective function and constraint condition building module is configured to: build an objective function with the profit maximization of the power sales company as the goal and build constraint conditions that meet the model; The model solving module is configured to solve the model based on linear programming, stochastic programming and robust optimization algorithms to obtain the optimal electricity purchasing strategy.

8. An electronic device, characterized in that: The method comprises a memory and a processor and computer instructions stored in the memory and executed on the processor, wherein when the computer instructions are executed by the processor, the method according to any one of claims 1 to 6 is completed.

9. A computer-readable storage medium, characterized in that: Used to store computer instructions, which, when executed by a processor, complete the method described in any one of claims 1 to 6.

10. A computer program product, characterized in that The invention comprises a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 6.

Images