Energy scheduling method and device considering dynamic carbon emission factor and computer equipment

Abstract

According to the energy scheduling method and device considering the dynamic carbon emission factor and the computer equipment provided by the invention, the recent new energy output data, the load data and the dynamic carbon emission factor set are obtained every time a scheduling time period is entered, and then the data are input into the preset fusion prediction network; therefore, the output prediction sequence and the load prediction sequence in the current scheduling time period can be determined based on the latest data, so that the output prediction sequence and the load prediction sequence can be subsequently determined based on the output prediction sequence, the load prediction sequence and the dynamic carbon emission factor set. The dual-objective optimization model constructed by taking minimization of the carbon emission and the power generation cost as an optimization objective and the constraint condition set thereof can ensure the adaptability and decision reliability of the dual-objective optimization model while considering the power generation cost and the environmental benefits. And finally, solving the dual-objective optimization model according to the constraint condition set, so that an optimal scheduling scheme obtained based on a solving result has relatively high scheduling effectiveness and reliability.

Description

Technical Field

[0001] The present application relates to the field of power grid dispatching technology, and in particular to an energy dispatching method, apparatus, and computer equipment that take into account dynamic carbon emission factors. Background Art

[0002] Currently, traditional energy dispatch methods are mostly based on deterministic optimization models. These methods played an important role in the past in centralized energy supply systems. However, with the widespread integration of distributed renewable energy, the operating environment of the power grid has become more complex and dynamic. Traditional dispatch methods struggle to effectively handle the randomness and volatility of distributed renewable energy output, resulting in a very limited contribution to the smooth operation of the power grid and ignoring the environmental benefits of grid operation. In summary, traditional dispatch methods struggle to improve the effectiveness of energy dispatch while balancing economic and environmental benefits. Summary of the Invention

[0003] The purpose of this application is to solve at least one of the above-mentioned technical defects, especially the technical defect that traditional scheduling methods in the existing technology are difficult to improve the effectiveness of energy scheduling while taking into account both economic and environmental benefits.

[0004] In a first aspect, the present application provides an energy scheduling method taking into account a dynamic carbon emission factor, the method comprising:

[0005] When entering each dispatch period, obtain new energy output data, load data and dynamic carbon emission factor set, and determine the preset fusion prediction network;

[0006] Inputting the new energy output data and the load data into the fusion prediction network respectively to obtain an output prediction sequence and a load prediction sequence within the current scheduling period;

[0007] Based on the output forecast sequence, the load forecast sequence and the dynamic carbon emission factor set, a dual-objective optimization model corresponding to the current scheduling period and its constraint condition set are constructed with minimization of carbon emissions and power generation costs as optimization objectives;

[0008] The dual-objective optimization model is solved according to the constraint condition set to obtain the scheduling sequence of the current scheduling period, and then the optimal scheduling solution at the current moment is generated according to the decision variable set corresponding to the first moment in the scheduling sequence.

[0009] In one embodiment, the fusion prediction network includes an LSTM model, an ARIMA model, and a Prophet model; the step of inputting the new energy output data and the load data into the fusion prediction network to obtain an output prediction sequence and a load prediction sequence within the current scheduling period includes:

[0010] Inputting the new energy output data and the load data into the LSTM model, the ARIMA model, and the Prophet model, respectively, to obtain a first prediction sequence, a second prediction sequence, and a third prediction sequence corresponding to the new energy output data, and a fourth prediction sequence, a fifth prediction sequence, and a sixth prediction sequence corresponding to the load data;

[0011] Determining weights of the LSTM model, the ARIMA model, and the Prophet model according to the data type of the new energy output data, so as to perform weighted fusion on the first prediction sequence, the second prediction sequence, and the third prediction sequence to obtain an output prediction sequence;

[0012] The weights of the LSTM model, the ARIMA model, and the Prophet model are determined according to the data type of the load data to perform weighted fusion on the fourth prediction sequence, the fifth prediction sequence, and the sixth prediction sequence to obtain a load forecast sequence.

[0013] In one embodiment, after obtaining the output forecast sequence and the load forecast sequence in the current scheduling period, the method further includes:

[0014] The output forecast sequence and the load forecast sequence are adjusted respectively by using a Kalman filtering method, so that the adjusted output forecast sequence and load forecast sequence are used as the final output forecast sequence and load forecast sequence in the current scheduling period.

[0015] In one embodiment, based on the output forecast sequence, the load forecast sequence, and the dynamic carbon emission factor set, a dual-objective optimization model corresponding to the current scheduling period and its constraint condition set are constructed with minimization of carbon emissions and power generation costs as optimization objectives, including:

[0016] Determining a power generation cost item based on the output forecast sequence and a preset cost coefficient set, and determining a carbon emission item based on the output forecast sequence and the dynamic carbon emission factor set;

[0017] Taking minimizing the sum of the power generation cost item and the carbon emission item as an optimization goal, a dual-objective optimization model is constructed based on the power generation cost item and the carbon emission item;

[0018] The power generation side constraints, energy storage side constraints, load side constraints and grid interaction constraints of the dual-objective optimization model are determined according to the load forecast sequence and the output forecast sequence to form a constraint condition set.

[0019] In one embodiment, the dual-objective optimization model is expressed as:

[0020]

[0021] Where, Indicates the start time of the current scheduling period, Indicates the duration of the current scheduling period. Indicates the power source exist The output power at the moment, Indicates the power source exist Dynamic carbon emission factor at each moment, 、 、 Indicates the power source The cost coefficient, Indicates the preset economic weight, Indicates the preset carbon emission weight, It represents the set of decision variables when the dual-objective optimization model achieves the optimization goal under the premise of satisfying the constraint condition set. Represents power generation resources including traditional power generation and distributed renewable energy power generation.

[0022] In one embodiment, solving the dual-objective optimization model according to the constraint condition set to obtain a scheduling sequence for the current scheduling period includes:

[0023] Identifying a nonlinear part in the dual-objective optimization model and simplifying the nonlinear part into a piecewise linear function to form a new dual-objective optimization model;

[0024] Determine high-carbon units and low-carbon units in the power grid according to the dynamic carbon emission factor set, set the low-carbon units as priority dispatch units, and determine the decision variable set corresponding to the high-carbon units as the decision variable range of the new dual-objective optimization model;

[0025] According to the decision variable range and the constraint condition set, the new dual-objective optimization model is solved and calculated using a solver to obtain the decision variable set of the dual-objective optimization model at each moment in the current scheduling period, and the decision variable set at each moment is used as a sequence element to generate a scheduling sequence for the current scheduling period.

[0026] In one embodiment, after generating the optimal scheduling solution at the current moment based on the decision variable set corresponding to the first moment in the scheduling sequence, the method further includes:

[0027] Execute the optimal scheduling plan and monitor the power grid operation status. If the power grid operation status does not reach the expected state corresponding to the optimal scheduling plan, adjust the fixed value parameters in the dual-objective optimization model to update the dual-objective optimization model.

[0028] In a second aspect, the present application provides an energy scheduling device that takes into account a dynamic carbon emission factor, the device comprising:

[0029] The data acquisition module is used to obtain new energy output data, load data and dynamic carbon emission factor sets when entering each scheduling period, and determine the preset fusion prediction network;

[0030] A sequence prediction module, configured to input the new energy output data and the load data into the fusion prediction network respectively, to obtain an output prediction sequence and a load prediction sequence within a current scheduling period;

[0031] a model construction module, configured to construct a dual-objective optimization model and its constraint condition set corresponding to the current scheduling period based on the output forecast sequence, the load forecast sequence, and the dynamic carbon emission factor set, with minimization of carbon emissions and power generation costs as optimization objectives;

[0032] The solution generation module is used to solve the dual-objective optimization model according to the constraint condition set, obtain the scheduling sequence of the current scheduling period, and generate the optimal scheduling solution at the current moment according to the decision variable set corresponding to the first moment in the scheduling sequence.

[0033] In a third aspect, the present application provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the energy scheduling method taking into account the dynamic carbon emission factor as described in any of the above embodiments.

[0034] In a fourth aspect, the present application provides a computer device, comprising: one or more processors, and a memory;

[0035] The memory stores computer-readable instructions, and when the one or more processors execute the computer-readable instructions, they perform the steps of the energy scheduling method taking into account the dynamic carbon emission factor as described in any of the above embodiments.

[0036] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0037] The energy scheduling method, device and computer equipment provided by the present application, which take into account dynamic carbon emission factors, obtains recent new energy output data, load data and dynamic carbon emission factor sets during the energy scheduling process each time a scheduling period is entered, and then the new energy output data and load data are respectively input into a preset fusion prediction network. In this way, the output forecast sequence and load forecast sequence in the current scheduling period can be determined based on the latest data, thereby improving the reliability of the prediction sequence. The dual-objective optimization model and its constraint condition set constructed based on the output forecast sequence, load forecast sequence and dynamic carbon emission factor set with carbon emissions and power generation cost minimization as the optimization objectives can ensure the adaptability and decision reliability of the dual-objective optimization model while taking into account power generation costs and environmental benefits. Finally, the dual-objective optimization model is solved according to the constraint condition set, so that the optimal scheduling solution obtained based on the solution result has high scheduling effectiveness and reliability. Moreover, when multiple scheduling periods are set in a rolling manner, the optimal scheduling solution can be regenerated when entering each scheduling period, and the scheduling solution can be dynamically adjusted to cope with the uncertainty of new energy power generation and load demand, so as to fully utilize resources and maximize the new energy consumption rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0039] Figure 1 A flow chart of an energy scheduling method taking into account a dynamic carbon emission factor provided in an embodiment of the present application;

[0040] Figure 2 A schematic diagram of a process for inputting renewable energy output data and load data into a fusion prediction network, provided in an embodiment of the present application;

[0041] Figure 3 A flow chart of a dual-objective optimization model and its constraint condition set corresponding to the current scheduling period, provided in an embodiment of the present application, with minimization of carbon emissions and power generation costs as optimization objectives;

[0042] Figure 4 A schematic diagram of a process for solving a dual-objective optimization model according to a set of constraints provided in an embodiment of the present application;

[0043] Figure 5 A schematic diagram of the structure of an energy scheduling device taking into account a dynamic carbon emission factor provided in an embodiment of the present application;

[0044] Figure 6 This is a diagram of the internal structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0046] In one embodiment, this application provides an energy scheduling method that takes into account dynamic carbon emission factors. The following embodiment illustrates this method as applied to a server. It is understood that this energy scheduling method that takes into account dynamic carbon emission factors can be implemented on a single server or a server cluster consisting of multiple servers, and this application does not impose specific limitations on this.

[0047] like Figure 1 As shown, the present application provides an energy scheduling method taking into account a dynamic carbon emission factor, the method comprising:

[0048] S101: When entering each scheduling period, obtain new energy output data, load data and dynamic carbon emission factor set, and determine the preset fusion prediction network.

[0049] In this step, when the user needs to schedule the energy in the power grid, the server can determine multiple scheduling periods through the preset scheduling period determination rules. Every time a scheduling moment is entered, the latest new energy output data, load data and dynamic carbon emission factor set are obtained through the sensor network and historical operation data in the power grid, and then the pre-trained fusion prediction network is determined.

[0050] Renewable energy output data refers to the actual power generation data of renewable energy sources in the power grid over a period of time. Load data refers to the power demand data of users over a period of time. The dynamic carbon emission factor set refers to the collection of carbon emission factor data of each power generation source in the power grid over a period of time. The dynamic carbon emission factor set can be obtained from the regional carbon emission factor database. The fusion forecasting network is used to predict renewable energy output and load demand over a period of time.

[0051] Furthermore, when obtaining new energy output data and load data, the obtained original output data and original load data can be normalized to serve as model input, and then the normalized original output data can be used as new energy output data, and the normalized original load data can be used as load data.

[0052] Specifically, the scheduling period determination rule can be set to a rolling determination. For example, if the scheduling period is set to 24 hours, and one of the scheduling periods is from 9:00 on the 8th to 9:00 on the 9th, and the interval is set to 2 hours, then the next scheduling period will be from 11:00 on the 8th to 11:00 on the 9th, and so on. Because an optimal scheduling plan is generated for each scheduling period, the rolling determination of scheduling periods can dynamically determine the scheduling plan to adapt to the uncertainty of renewable energy generation and load demand, thereby ensuring the smooth operation of the power grid.

[0053] S102: Inputting the new energy output data and load data into the fusion prediction network respectively to obtain the output prediction sequence and load prediction sequence in the current scheduling period.

[0054] The output forecast sequence is a sequence of output data at each moment in the current dispatch period, and the load forecast sequence is a sequence of load data at each moment in the current dispatch period. It is understood that the output data at each moment includes the output power of each power generation source in the power grid at that moment.

[0055] In this step, since the fusion prediction network can predict the new energy output and load demand in the future period based on the output data and load data respectively, the new energy output data and load data are input into the fusion prediction network, so that the fusion prediction network can predict the new energy output and load demand through multiple base models and use historical new energy output data and load data, and can determine the output prediction sequence and load prediction sequence in the current scheduling period.

[0056] Specifically, the fusion prediction network includes multiple base models. When the fusion prediction network processes different data, the weights of each base model can be determined according to factors such as the complexity and characteristics of the processed data, thereby ensuring that the fusion prediction network can maintain high reliability and accuracy when processing different data, and improving the generalization ability of the fusion prediction network.

[0057] S103: Based on the output forecast sequence, the load forecast sequence and the dynamic carbon emission factor set, a dual-objective optimization model corresponding to the current scheduling period and its constraint condition set are constructed with minimization of carbon emissions and power generation costs as optimization goals.

[0058] The constraint condition set includes the constraint conditions involved in multiple ends, such as: power generation side, load side, etc.

[0059] In this step, after determining the output forecast sequence, load forecast sequence, and dynamic carbon emission factor set, a dual-objective optimization model corresponding to the current scheduling moment and its corresponding set of constraints are constructed based on these sequences, with minimizing carbon emissions and power generation costs as the optimization objectives. This dual-objective optimization model aims to minimize both carbon emissions and power generation costs, thus balancing economic and environmental benefits, maximizing environmental benefits while controlling power generation costs.

[0060] S104: Solve the dual-objective optimization model according to the constraint condition set to obtain the scheduling sequence of the current scheduling period, and then generate the optimal scheduling plan at the current moment according to the decision variable set corresponding to the first moment in the scheduling sequence.

[0061] The decision variable set refers to the feasible solution of the bi-objective optimization model, which includes multiple decision variables.

[0062] In this step, the feasible solution of the dual-objective optimization model at the minimum is solved within the scope of the constraint set, that is, the decision variable set, and the feasible solution of the dual-objective optimization model at each moment in the current scheduling period is obtained to form the scheduling sequence corresponding to the current scheduling period. Then, the optimal scheduling plan at the current moment is generated based on the decision variable set corresponding to the first moment in the scheduling sequence.

[0063] Specifically, the first moment in the scheduling sequence is the first moment in the current scheduling period. In other words, the first moment in the scheduling sequence is the moment corresponding to entering the current scheduling period. Therefore, the decision variable set corresponding to the first moment in the scheduling sequence is used as the data for generating the optimal scheduling plan, thereby ensuring the real-time and reliability of the scheduling plan to the greatest extent.

[0064] After determining the optimal dispatch plan, it can be executed to dispatch grid resources. Specifically, the optimal dispatch plan can be mainly divided into four aspects: power generation side, energy storage measurement, load side and grid interaction. For example: issuing generator output instructions and adjusting the operating parameters of equipment such as gas turbines and photovoltaic inverters; controlling the charge and discharge power of the energy storage system to ensure that the SOC is within a safe range; sending dispatch signals to adjustable loads through demand-side management (DSM) (such as delayed start-up and reduced power operation); adjusting the grid interaction power to ensure that the bus voltage and line flow are within safety limits.

[0065] In one example, when determining the decision variable set corresponding to the first moment in the scheduling sequence, a preset scheduling plan template can be obtained, and then the decision variable set can be filled into the scheduling plan template to obtain the optimal scheduling plan. Through the unified template format, subsequent computer devices can understand and execute the optimal scheduling plan.

[0066] In the above embodiment, during the energy scheduling process, upon entering each scheduling period, recent renewable energy output data, load data, and a dynamic carbon emission factor set are obtained. The renewable energy output data and load data are then input into a pre-set fusion prediction network. This allows the output forecast sequence and load forecast sequence for the current scheduling period to be determined based on the latest data, improving the reliability of the forecast sequence. This allows the subsequent dual-objective optimization model and its constraint condition set, constructed with minimizing carbon emissions and power generation costs as optimization objectives based on the output forecast sequence, load forecast sequence, and dynamic carbon emission factor set, to ensure the adaptability and decision reliability of the dual-objective optimization model while taking into account both power generation costs and environmental benefits. Finally, the dual-objective optimization model is solved based on the constraint condition set, ensuring that the optimal scheduling solution obtained based on the solution has high scheduling effectiveness and reliability. Furthermore, when multiple scheduling periods are set in a rolling manner, the optimal scheduling solution can be regenerated upon entering each scheduling period, dynamically adjusting the scheduling solution to address the uncertainty of renewable energy generation and load demand, thereby fully utilizing resources and maximizing the renewable energy absorption rate.

[0067] like Figure 2 As shown, in one embodiment, the fusion prediction network includes an LSTM model, an ARIMA model, and a Prophet model; the new energy output data and load data are respectively input into the fusion prediction network to obtain the output prediction sequence and load prediction sequence in the current scheduling period, including:

[0068] S201: Input the new energy output data and load data into the LSTM model, ARIMA model and Prophet model respectively to obtain the first prediction sequence, second prediction sequence and third prediction sequence corresponding to the new energy output data and the fourth prediction sequence, fifth prediction sequence and sixth prediction sequence corresponding to the load data.

[0069] The LSTM (Long Short-Term Memory) model is a recurrent neural network that can be used to capture long-term dependencies in sequence data. The ARIMA (Autoregressive Integrated Moving Average) model is a statistical time series model that can be used for modeling and forecasting stationary time series. It captures trends and seasonal changes in time series data by combining three main components: autoregression, differencing, and smoothed moving average. The Prophet model is a time series forecasting model primarily used to process time series data with strong seasonality and trends.

[0070] In this step, the new energy output data and load data are respectively input into the fusion prediction network, and the LSTM model in the fusion prediction network is used to perform output prediction and load prediction, and the first prediction sequence corresponding to the new energy output data and the fourth prediction sequence corresponding to the load data are obtained respectively. Then, the ARIMA model in the fusion prediction network is used to perform output prediction and load prediction, and the second prediction sequence corresponding to the new energy output data and the fifth prediction sequence corresponding to the load data are obtained respectively. Finally, the Prophet model in the fusion prediction network is used to perform output prediction and load prediction, and the third prediction sequence corresponding to the new energy output data and the sixth prediction sequence corresponding to the load data are obtained respectively.

[0071] Specifically, when using a fusion prediction network composed of LSTM model, ARIMA model and Prophet model to predict output data and load data, the advantages of each model can be fully utilized to improve the accuracy of the prediction, and it can also offset the limitations of a single model to a certain extent, providing higher prediction accuracy and better generalization ability.

[0072] S202: Determine the weights of the LSTM model, the ARIMA model, and the Prophet model according to the data type of the new energy output data, so as to perform weighted fusion on the first prediction sequence, the second prediction sequence, and the third prediction sequence to obtain an output prediction sequence.

[0073] S203: Determine the weights of the LSTM model, the ARIMA model, and the Prophet model according to the data type of the load data, so as to perform weighted fusion on the fourth prediction sequence, the fifth prediction sequence, and the sixth prediction sequence to obtain a load prediction sequence.

[0074] Specifically, in the application scenarios of renewable energy output and load forecasting, data types typically include time series data (such as historical renewable energy output data and meteorological data) and periodic data (such as seasonal changes and diurnal alternation). To fully utilize these data features, the advantages of the LSTM model, ARIMA model, and Prophet model can be combined to generate the final output and load forecast series through weighted fusion.

[0075] In this embodiment, the weights of the LSTM model, ARIMA model, and Prophet model are determined based on the type and characteristics of the new energy output data or load data. For example, if the data has obvious nonlinear characteristics and long-term dependencies, the weight of the LSTM model can be set higher; if the data shows strong periodicity and seasonality, the weight of the Prophet model can be appropriately increased; and when the data is relatively stable and has a certain linear trend, the weight of the ARIMA model can also be appropriately increased. After determining the appropriate weights, the first prediction sequence, the second prediction sequence, and the third prediction sequence or the fourth prediction sequence, the fifth prediction sequence, and the sixth prediction sequence are weighted and fused based on the weight of each model to obtain the output forecast sequence or load forecast data.

[0076] For example, consider photovoltaic power output forecasting, which is influenced by weather conditions (such as sunlight intensity and cloud cover) and the diurnal cycle. In this case, the LSTM model can leverage its ability to capture nonlinear characteristics to handle complex meteorological data; the Prophet model can leverage its strengths in modeling seasonal and cyclical variations to address diurnal and seasonal variations; and the ARIMA model can be used to address short-term linear trends in the data. By assigning appropriate weights to each model (for example, LSTM weight of 0.4, Prophet weight of 0.3, and ARIMA weight of 0.3), and combining the prediction results of the three models, a more accurate photovoltaic power output forecast series can be obtained. Similarly, in wind power output forecasting, since wind power is significantly affected by wind speed variations, the LSTM model can better capture the dynamic changes in wind speed, while the Prophet model can handle the seasonal characteristics of wind power, and the ARIMA model is used to address short-term trends.

[0077] In another example, consider load forecasting in residential areas. Load is typically low during the day and peaks at night, exhibiting significant diurnal periodicity. Load is also affected by seasonal variations (such as increased air conditioning use in the summer) and weather conditions (such as increased heating use due to cold weather). In this case, a higher weight (such as 0.4) can be assigned to the Prophet model, as it effectively handles diurnal and seasonal variations. A moderate weight (such as 0.3) can be assigned to the LSTM model to capture the impact of weather changes on load. The remaining weight (such as 0.3) can be assigned to the ARIMA model to account for short-term linear trends. Finally, a more accurate load forecast sequence can be obtained through weighted fusion.

[0078] It can be understood that the weights of different models in the fusion prediction network are dynamically determined according to the type and characteristics of the input data, so that the fusion prediction network can better adapt to different types of input data, give full play to the advantages of each model, and improve the prediction accuracy.

[0079] In one embodiment, after obtaining the output forecast sequence and the load forecast sequence within the current scheduling period, the energy scheduling method taking into account the dynamic carbon emission factor further includes:

[0080] The Kalman filter method is used to adjust the output forecast sequence and the load forecast sequence respectively, so that the adjusted output forecast sequence and the load forecast sequence are used as the final output forecast sequence and the load forecast sequence in the current scheduling period.

[0081] In this embodiment, Kalman filtering can be used to perform error correction on the output forecast sequence and the load forecast sequence, thereby further ensuring the accuracy and reliability of the forecast data.

[0082] like Figure 3 As shown, in one embodiment, based on the output forecast sequence, the load forecast sequence and the dynamic carbon emission factor set, a dual-objective optimization model corresponding to the current scheduling period and its constraint condition set are constructed with minimization of carbon emissions and power generation costs as the optimization objectives, including:

[0083] S301: Determine a power generation cost item based on an output forecast sequence and a preset cost coefficient set, and determine a carbon emission item based on an output forecast sequence and a dynamic carbon emission factor set.

[0084] The power generation cost item refers to a data item representing the power generation cost, and the carbon emission item refers to a data item representing the carbon emission. The cost coefficient set includes multiple cost coefficients corresponding to each power generation source in the power grid.

[0085] In this step, the data item representing the power generation cost, i.e., the power generation cost item, is determined based on the output forecast sequence and the preset cost coefficient set, and then the data item representing the carbon emissions, i.e., the carbon emissions item, is determined based on the output forecast sequence and the dynamic carbon emission factor set.

[0086] S302: Taking minimizing the sum of the power generation cost item and the carbon emission item as the optimization goal, a dual-objective optimization model is constructed based on the power generation cost item and the carbon emission item.

[0087] In this step, we obtain the preset economic and carbon emission weights, and then construct a dual-objective optimization model by combining the power generation cost and carbon emission items with the optimization objective. This allows us to obtain a feasible solution that balances economic and environmental benefits.

[0088] S303: Determine the generation side constraints, energy storage side constraints, load side constraints, and grid interaction constraints of the dual-objective optimization model according to the load forecast sequence and the output forecast sequence to form a constraint condition set.

[0089] Among them, generation-side constraints refer to constraints related to power generation, including the output range of power generation equipment, ramp rate, unit start and stop status, power balance, etc. Energy storage-side constraints refer to constraints related to energy storage equipment, including its charge and discharge power limits, capacity limits, charge and discharge efficiency, and SOC status. Load-side constraints refer to constraints related to user loads, including the degree to which load demand can be met, restrictions on interruptible loads, and the amount of adjustable reduction and transfer. Grid interaction constraints refer to constraints related to grid power exchange, including the grid's exchange power.

[0090] In this step, the constraints of the dual-objective optimization model in terms of power generation, energy storage, load, and grid interaction are determined based on the load forecast sequence, output forecast sequence, and relevant parameters of the grid dispatch, so as to obtain power generation constraints, energy storage constraints, load constraints, and grid interaction constraints, thus forming a constraint condition set for the dual-objective optimization model.

[0091] In an example, the power balance constraint in the generation side constraint can be expressed as:

[0092]

[0093] Where, Represents power generation resources including traditional power generation and distributed new energy power generation, Indicates the power source exist The output power at the moment, Indicates that the energy storage device is Storage power at the moment, For the load side Load power at the moment.

[0094] Specifically, by integrating these constraints, the resulting constraint set provides clear boundary conditions for the dual-objective optimization model, enabling it to find the optimal solution in a realistic operating environment. The dual-objective optimization model, constructed with minimizing carbon emissions and power generation costs as optimization objectives, and its constraint set ensures both the adaptability and decision-making reliability of the dual-objective optimization model while balancing power generation costs and environmental benefits.

[0095] In one embodiment, the dual-objective optimization model is expressed as:

[0096]

[0097] Where, Indicates the start time of the current scheduling period, Indicates the duration of the current scheduling period. Indicates the power source exist The output power at the moment, Indicates the power source exist Dynamic carbon emission factor at each moment, 、 、 Indicates the power source The cost coefficient, Indicates the preset economic weight, Indicates the preset carbon emission weight, It represents the set of decision variables when the dual-objective optimization model achieves the optimization goal under the premise of satisfying the constraint condition set. Represents power generation resources including traditional power generation and distributed renewable energy power generation.

[0098] like Figure 4 As shown, in one embodiment, the dual-objective optimization model is solved according to the constraint condition set to obtain the scheduling sequence for the current scheduling period, including:

[0099] S401: Identify a nonlinear part in a dual-objective optimization model and simplify the nonlinear part into a piecewise linear function to form a new dual-objective optimization model.

[0100] In this step, the nonlinear part of the dual-objective optimization model is identified, and then the identified nonlinear part is simplified into a piecewise linear function to obtain a new dual-objective optimization model. In one example, the identified nonlinear part can be simplified into a piecewise linear function according to the following expression:

[0101]

[0102] Where, represents the nonlinear part in the dual-objective optimization model, Indicates the number of segments, represents the output power of the power source i in the kth power interval at time t, 、 Indicates the piecewise linear coefficient corresponding to the k-th power interval.

[0103] This process can transform complex nonlinear problems into linear problems that are easier to solve, thereby reducing the complexity and computational cost of the problem. Through piecewise linearization, mature linear programming tools and algorithms can be used to efficiently solve optimization problems while maintaining the approximate accuracy of the original problem.

[0104] S402: Determine the high-carbon units and low-carbon units in the power grid based on the dynamic carbon emission factor set, set the low-carbon units as priority scheduling units, and determine the decision variable set corresponding to the high-carbon units as the decision variable range of the new dual-objective optimization model.

[0105] In this step, the carbon emissions of each power source in the grid are determined using a dynamic carbon emission factor set. The power sources in the grid are then divided to identify high-carbon and low-carbon units. The low-carbon units are then prioritized for scheduling, and the decision variables for these prioritized units are excluded from the decision variable range. At this point, only the decision variable set corresponding to the high-carbon units is required to be included in the decision variable range of the new dual-objective optimization model. By narrowing the decision variable range of the dual-objective optimization model, the problem dimension is reduced, thereby reducing solution complexity, conserving computing resources, and improving solution efficiency.

[0106] Furthermore, when classifying power generation sources within a power grid, the classification criteria can be determined based on whether the dynamic carbon emission factor is above a preset threshold. Generating sources with dynamic carbon emission factors above the threshold can be classified as high-carbon units, while sources with dynamic emission factors below the threshold can be classified as low-carbon units. It is understood that the preset threshold is an empirical value and can be modified based on actual conditions and needs.

[0107] S403: Based on the decision variable range and constraint condition set, the new dual-objective optimization model is solved and calculated using the solver to obtain the decision variable set of the dual-objective optimization model at each moment in the current scheduling period, and the decision variable set at each moment is used as a sequence element to generate the scheduling sequence of the current scheduling period.

[0108] Each sequence element in the scheduling sequence contains a set of decision variables.

[0109] In this step, based on the range of decision variables and the set of constraints, the simplified dual-objective optimization model is solved using a solver. This way, the feasible solution of the dual-objective optimization model corresponding to each moment in the current scheduling period can be obtained, that is, the decision variable set. The decision variable set at each moment is then used as a sequence element to form the scheduling sequence of the current scheduling period.

[0110] Specifically, by narrowing the scope of decision variables and simplifying the dual-objective optimization model, the solution efficiency can be optimized, computing resources can be saved, and the real-time response capability of dynamic scheduling solution generation can be improved.

[0111] In one embodiment, after generating the optimal scheduling solution at the current moment based on the decision variable set corresponding to the first moment in the scheduling sequence, the energy scheduling method taking into account the dynamic carbon emission factor further includes:

[0112] The optimal dispatching plan is executed and the grid operation status is monitored. If the grid operation status does not reach the expected state corresponding to the optimal dispatching plan, the fixed value parameters in the dual-objective optimization model are adjusted to update the dual-objective optimization model.

[0113] Specifically, the fixed-value parameters in the dual-objective optimization model refer to some preset parameters, such as cost coefficient, economic weight, carbon emission weight, etc.

[0114] In this embodiment, after executing the optimal scheduling plan, the grid operating status can be monitored to analyze whether the optimal scheduling plan has achieved the expected scheduling effect. If it is determined that the optimal scheduling plan has not achieved the expected scheduling effect, the fixed value parameters in the dual-objective optimization model can be adjusted to optimize and update the dual-objective optimization model. This feedback closed-loop optimization method can continuously optimize the dual-objective optimization model when the scheduling plan does not meet expectations, so that the subsequent generation of the optimal scheduling plan has higher scheduling effectiveness.

[0115] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0116] The energy scheduling device taking into account the dynamic carbon emission factor provided in an embodiment of the present application is described below. The energy scheduling device taking into account the dynamic carbon emission factor described below and the energy scheduling method taking into account the dynamic carbon emission factor described above can be referenced to each other.

[0117] like Figure 5 As shown, the present application provides an energy scheduling device 500 that takes into account a dynamic carbon emission factor, the device comprising:

[0118] The data acquisition module 501 is used to obtain new energy output data, load data and dynamic carbon emission factor set when entering each scheduling period, and determine the preset fusion prediction network;

[0119] The sequence prediction module 502 is used to input the new energy output data and load data into the fusion prediction network to obtain the output prediction sequence and load prediction sequence in the current scheduling period;

[0120] Model construction module 503 is used to construct a dual-objective optimization model and its constraint condition set corresponding to the current scheduling period based on the output forecast sequence, the load forecast sequence and the dynamic carbon emission factor set, with minimization of carbon emissions and power generation costs as the optimization objectives;

[0121] The solution generation module 504 is used to solve the dual-objective optimization model according to the constraint condition set, obtain the scheduling sequence of the current scheduling period, and generate the optimal scheduling solution at the current moment according to the decision variable set corresponding to the first moment in the scheduling sequence.

[0122] In the above embodiment, during the energy scheduling process, upon entering each scheduling period, recent renewable energy output data, load data, and a dynamic carbon emission factor set are obtained. The renewable energy output data and load data are then input into a pre-set fusion prediction network. This allows the output forecast sequence and load forecast sequence for the current scheduling period to be determined based on the latest data, improving the reliability of the forecast sequence. This allows the subsequent dual-objective optimization model and its constraint condition set, constructed with minimizing carbon emissions and power generation costs as optimization objectives based on the output forecast sequence, load forecast sequence, and dynamic carbon emission factor set, to ensure the adaptability and decision reliability of the dual-objective optimization model while taking into account both power generation costs and environmental benefits. Finally, the dual-objective optimization model is solved based on the constraint condition set, ensuring that the optimal scheduling solution obtained based on the solution has high scheduling effectiveness and reliability. Furthermore, when multiple scheduling periods are set in a rolling manner, the optimal scheduling solution can be regenerated upon entering each scheduling period, dynamically adjusting the scheduling solution to address the uncertainty of renewable energy generation and load demand, thereby fully utilizing resources and maximizing the renewable energy absorption rate.

[0123] In one embodiment, the fusion prediction network includes an LSTM model, an ARIMA model, and a Prophet model; the sequence prediction module includes:

[0124] The sequence prediction submodule is used to input the renewable energy output data and load data into the LSTM model, ARIMA model, and Prophet model respectively to obtain the first prediction sequence, the second prediction sequence, and the third prediction sequence corresponding to the renewable energy output data, and the fourth prediction sequence, the fifth prediction sequence, and the sixth prediction sequence corresponding to the load data;

[0125] The first fusion submodule is used to determine the weights of the LSTM model, the ARIMA model, and the Prophet model according to the data type of the new energy output data, so as to perform weighted fusion on the first prediction sequence, the second prediction sequence, and the third prediction sequence to obtain the output prediction sequence;

[0126] The second fusion submodule is used to determine the weights of the LSTM model, the ARIMA model, and the Prophet model according to the data type of the load data, so as to perform weighted fusion on the fourth prediction sequence, the fifth prediction sequence, and the sixth prediction sequence to obtain a load forecast sequence.

[0127] In one embodiment, the energy scheduling device taking into account the dynamic carbon emission factor further includes:

[0128] The sequence filtering module is used to adjust the output forecast sequence and the load forecast sequence respectively by using the Kalman filtering method, so as to use the adjusted output forecast sequence and the load forecast sequence as the final output forecast sequence and the load forecast sequence in the current scheduling period.

[0129] In one embodiment, the model building module includes:

[0130] A data item determination submodule, configured to determine a power generation cost item based on an output forecast sequence and a preset cost coefficient set, and to determine a carbon emission item based on an output forecast sequence and a dynamic carbon emission factor set;

[0131] The model construction submodule is used to construct a dual-objective optimization model based on the power generation cost item and the carbon emission item, taking minimizing the sum of the power generation cost item and the carbon emission item as the optimization objective;

[0132] The constraint formation submodule is used to determine the generation side constraints, energy storage side constraints, load side constraints and grid interaction constraints of the dual-objective optimization model according to the load forecast sequence and the output forecast sequence to form a constraint condition set.

[0133] In one embodiment, the solution generation module includes:

[0134] The model simplification submodule is used to identify the nonlinear part in the dual-objective optimization model and simplify the nonlinear part into a piecewise linear function to form a new dual-objective optimization model;

[0135] The scope determination submodule is used to determine the high-carbon units and low-carbon units in the power grid based on the dynamic carbon emission factor set, set the low-carbon units as the priority scheduling units, and determine the decision variable set corresponding to the high-carbon units as the decision variable range of the new dual-objective optimization model;

[0136] The model solving submodule is used to solve the new dual-objective optimization model using the solver according to the range of decision variables and the set of constraints, obtain the decision variable set of the dual-objective optimization model at each moment in the current scheduling period, and use the decision variable set at each moment as a sequence element to generate the scheduling sequence of the current scheduling period.

[0137] In one embodiment, the energy scheduling device taking into account the dynamic carbon emission factor further includes:

[0138] The parameter adjustment module is used to execute the optimal scheduling plan and monitor the operating status of the power grid. If the operating status of the power grid does not reach the expected state corresponding to the optimal scheduling plan, the fixed value parameters in the dual-objective optimization model are adjusted to update the dual-objective optimization model.

[0139] The division of the various modules in the above-mentioned energy scheduling device taking into account the dynamic carbon emission factor is only for illustration. In other embodiments, the energy scheduling device taking into account the dynamic carbon emission factor can be divided into different modules as needed to complete all or part of the functions of the above-mentioned energy scheduling device taking into account the dynamic carbon emission factor. The various modules in the above-mentioned energy scheduling device taking into account the dynamic carbon emission factor can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0140] In one embodiment, the present application also provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the energy scheduling method taking into account the dynamic carbon emission factor as described in any of the above embodiments.

[0141] In one embodiment, the present application also provides a computer device having computer-readable instructions stored therein. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the energy scheduling method taking into account the dynamic carbon emission factor as described in any of the above embodiments.

[0142] Schematically, as Figure 6 As shown, Figure 6 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of the present application. The computer device 600 can be provided as a server. Figure 6 Computer device 600 includes a processing component 602, which further includes one or more processors, and memory resources represented by memory 601 for storing instructions executable by processing component 602, such as application programs. The application programs stored in memory 601 may include one or more modules, each corresponding to a set of instructions. Furthermore, processing component 602 is configured to execute the instructions to perform the energy scheduling method that takes into account a dynamic carbon emission factor according to any of the above-described embodiments.

[0143] The computer device 600 may further include a power supply component 603 configured to perform power management of the computer device 600, a wired or wireless network interface 604 configured to connect the computer device 600 to a network, and an input / output (I / O) interface 605. The computer device 600 may operate based on an operating system stored in the memory 601, such as Windows Server™, Mac OS X™, Unix™, Linux™, Free BSD™, or the like.

[0144] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0145] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment comprising a series of elements not only include those elements, but also include other elements not clearly listed, or also include elements inherent to such process, method, article or equipment. In the absence of more restrictions, the elements limited by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or equipment comprising the elements. Herein, the singular forms "one", "an" and "said / the" may also include plural forms, unless the context clearly indicates otherwise. It should also be understood that the terms "include / comprise" or "have" etc. specify the existence of stated features, wholes, steps, operations, components, parts or combinations thereof, but do not exclude the possibility of the existence or addition of one or more other features, wholes, steps, operations, components, parts or combinations thereof. At the same time, the term "and / or" used in this specification includes any and all combinations of the relevant listed items.

[0146] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referenced to each other.

[0147] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An energy scheduling method taking into account dynamic carbon emission factors, characterized in that: The method comprises: When entering each dispatch period, obtain new energy output data, load data and dynamic carbon emission factor set, and determine the preset fusion prediction network; Inputting the new energy output data and the load data into the fusion prediction network respectively to obtain an output prediction sequence and a load prediction sequence within the current scheduling period; Based on the output forecast sequence, the load forecast sequence and the dynamic carbon emission factor set, a dual-objective optimization model corresponding to the current scheduling period and its constraint condition set are constructed with minimization of carbon emissions and power generation costs as optimization objectives; The dual-objective optimization model is solved according to the constraint condition set to obtain the scheduling sequence of the current scheduling period, and then the optimal scheduling solution at the current moment is generated according to the decision variable set corresponding to the first moment in the scheduling sequence.

2. The energy scheduling method taking into account the dynamic carbon emission factor according to claim 1 is characterized in that: The fusion prediction network includes an LSTM model, an ARIMA model, and a Prophet model; the new energy output data and the load data are respectively input into the fusion prediction network to obtain an output prediction sequence and a load prediction sequence within the current scheduling period, including: Inputting the new energy output data and the load data into the LSTM model, the ARIMA model, and the Prophet model, respectively, to obtain a first prediction sequence, a second prediction sequence, and a third prediction sequence corresponding to the new energy output data, and a fourth prediction sequence, a fifth prediction sequence, and a sixth prediction sequence corresponding to the load data; Determining weights of the LSTM model, the ARIMA model, and the Prophet model according to the data type of the new energy output data, so as to perform weighted fusion on the first prediction sequence, the second prediction sequence, and the third prediction sequence to obtain an output prediction sequence; The weights of the LSTM model, the ARIMA model, and the Prophet model are determined according to the data type of the load data to perform weighted fusion on the fourth prediction sequence, the fifth prediction sequence, and the sixth prediction sequence to obtain a load forecast sequence.

3. The energy scheduling method taking into account the dynamic carbon emission factor according to claim 1 or 2, characterized in that: After obtaining the output forecast sequence and the load forecast sequence within the current scheduling period, the method further includes: The output forecast sequence and the load forecast sequence are adjusted respectively by using a Kalman filtering method, so that the adjusted output forecast sequence and load forecast sequence are used as the final output forecast sequence and load forecast sequence in the current scheduling period.

4. The energy scheduling method taking into account the dynamic carbon emission factor according to claim 1 is characterized in that: The dual-objective optimization model corresponding to the current scheduling period and its constraint condition set are constructed based on the output forecast sequence, the load forecast sequence and the dynamic carbon emission factor set, with minimization of carbon emissions and power generation costs as optimization objectives, including: Determining a power generation cost item based on the output forecast sequence and a preset cost coefficient set, and determining a carbon emission item based on the output forecast sequence and the dynamic carbon emission factor set; Taking minimizing the sum of the power generation cost item and the carbon emission item as an optimization goal, a dual-objective optimization model is constructed based on the power generation cost item and the carbon emission item; The power generation side constraints, energy storage side constraints, load side constraints and grid interaction constraints of the dual-objective optimization model are determined according to the load forecast sequence and the output forecast sequence to form a constraint condition set.

5. The energy scheduling method taking into account the dynamic carbon emission factor according to claim 1 or 4, characterized in that: The dual-objective optimization model is expressed as: Where, Indicates the start time of the current scheduling period, Indicates the duration of the current scheduling period. Indicates the power source exist The output power at the moment, Indicates the power source exist Dynamic carbon emission factor at each moment, 、 、 Indicates the power source The cost coefficient, Indicates the preset economic weight, Indicates the preset carbon emission weight, It represents the set of decision variables when the dual-objective optimization model achieves the optimization goal under the premise of satisfying the constraint condition set. Represents power generation resources including traditional power generation and distributed renewable energy power generation.

6. The energy scheduling method taking into account the dynamic carbon emission factor according to claim 1 is characterized in that: Solving the dual-objective optimization model according to the constraint condition set to obtain a scheduling sequence for the current scheduling period includes: Identifying a nonlinear part in the dual-objective optimization model and simplifying the nonlinear part into a piecewise linear function to form a new dual-objective optimization model; Determine high-carbon units and low-carbon units in the power grid according to the dynamic carbon emission factor set, set the low-carbon units as priority dispatch units, and determine the decision variable set corresponding to the high-carbon units as the decision variable range of the new dual-objective optimization model; According to the decision variable range and the constraint condition set, the new dual-objective optimization model is solved and calculated using a solver to obtain the decision variable set of the dual-objective optimization model at each moment in the current scheduling period, and the decision variable set at each moment is used as a sequence element to generate a scheduling sequence for the current scheduling period.

7. The energy scheduling method taking into account the dynamic carbon emission factor according to claim 1 is characterized in that: After generating the optimal scheduling solution at the current moment according to the decision variable set corresponding to the first moment in the scheduling sequence, the method further includes: Execute the optimal scheduling plan and monitor the power grid operation status. If the power grid operation status does not reach the expected state corresponding to the optimal scheduling plan, adjust the fixed value parameters in the dual-objective optimization model to update the dual-objective optimization model.

8. An energy scheduling device taking into account dynamic carbon emission factors, characterized in that: The device comprises: The data acquisition module is used to obtain new energy output data, load data and dynamic carbon emission factor sets when entering each scheduling period, and determine the preset fusion prediction network; A sequence prediction module, configured to input the new energy output data and the load data into the fusion prediction network respectively, to obtain an output prediction sequence and a load prediction sequence within a current scheduling period; a model construction module, configured to construct a dual-objective optimization model and its constraint condition set corresponding to the current scheduling period based on the output forecast sequence, the load forecast sequence, and the dynamic carbon emission factor set, with minimization of carbon emissions and power generation costs as optimization objectives; The solution generation module is used to solve the dual-objective optimization model according to the constraint condition set, obtain the scheduling sequence of the current scheduling period, and generate the optimal scheduling solution at the current moment according to the decision variable set corresponding to the first moment in the scheduling sequence.

9. A storage medium, characterized in that: The storage medium stores computer-readable instructions, which, when executed by one or more processors, enable the one or more processors to execute the steps of the energy scheduling method taking into account the dynamic carbon emission factor as described in any one of claims 1 to 7.

10. A computer device, characterized in that: include: one or more processors, and memory; The memory stores computer-readable instructions, which, when executed by the one or more processors, execute the steps of the energy scheduling method taking into account the dynamic carbon emission factor as described in any one of claims 1 to 7.

Images