Data processing method and storage medium
By training electricity price data, photovoltaic output data and load data, the minimum operating cost of the power system is predicted, and combined with the cost optimization of the green certificate, the problem of failure of the renewable energy portfolio standard mechanism is solved, and the efficient operation of the power system is achieved and the effective use of green electricity.
Patent Information
- Application Number
- CN202310736336.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-06-20
- Publication Date
- 2024-12-20
AI Technical Summary
The current technology of tradable green certificates are restricted nationwide and cannot change the existing power contracts and physical power transmission of enterprises, resulting in an increase in thermal power auxiliary service costs when green power penetration increases, which may lead to renewable energy portfolio standards. The mechanism is invalid.
By using electricity price data, photovoltaic output data and load data as training samples, the energy processing model is trained to predict the minimum operating cost of the power system, and combined with the purchase and sale costs of the green certificate, the operating strategy of the power system is optimized.
It realizes that the minimum operating costs of the power system are accurately predicted and optimized under the premise of ensuring the operation of the power system, reduces the operating costs of thermal power units, improves the utilization efficiency of green power, and delays the increase in the cost of thermal power auxiliary services.
Smart Images
Figure CN119167078A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of data processing technology, and in particular to a data processing method and a storage medium. Background Art
[0002] China's renewable energy quota system is being piloted, and the derived green certificate trading market is still in its initial stage. A green certificate trading market suitable for national conditions has not yet been formed. In order to smoothly implement the renewable energy quota system and ensure the normal operation of the green certificate trading market, it is necessary to design scientific and reasonable market parameters and simulate their operating effects. In the related technology, the tradable green certificate (TGC) in the prior art is subject to non-bundled sales, which limits TGC to be sold separately from potential energy and can be used nationwide. In this case, although such TGCs provide a flexible way to support renewable energy development, they cannot change the company's existing electricity contracts and physical power delivery.
[0003] The clearing price of the spot market is based on the economic dispatch in the power system. Green electricity (GE) enjoys priority dispatch rights due to its zero marginal cost. However, no matter how high the Renewable Portfolio Standard (RPS) quota ratio is set, the reduction of GE is inevitable when the GE penetration rate increases. As the GE ratio increases, the RPS mechanism may gradually become ineffective: the additional consumption of GE will greatly increase the ancillary service costs of thermal power and occupy the share of thermal power, making it impossible to further consume GE in cost-oriented economic dispatch. Summary of the invention
[0004] The present disclosure provides a data processing method and a storage medium.
[0005] According to a first aspect of an embodiment of the present disclosure, there is provided a data processing method, including:
[0006] The energy processing model is trained using electricity price data, photovoltaic output data, and load data as training samples;
[0007] During the iterative process of training the energy processing model, when the operating cost of the next iteration output of the energy processing model is less than the operating cost of the previous iteration output of the energy processing model, and the difference between the operating cost of the previous iteration output and the operating cost of the next iteration output is less than a first preset difference, the learnable parameters of the next iteration of the energy processing model are used as the target learnable parameters of the energy processing model to obtain a trained energy processing model, and the trained energy processing model is used to predict the minimum operating cost of the power system.
[0008] Optionally, when the energy processing model is an energy management scheduling model, the operating cost of the power system is the sum of the power cost of the thermal power units, the green certificate cost, the transaction cost of the public power grid, the shared reserve capacity cost and the load transfer cost.
[0009] Optionally, the method further comprises:
[0010] When the actual amount of green electricity consumed is less than the preset amount of green electricity, the green electricity consumption, the unit price of the first green certificate purchased, and the transaction penalty coefficient, the green certificate cost is obtained;
[0011] When the green electricity consumption is greater than or equal to the preset green electricity amount, the green certificate cost is obtained according to the green electricity consumption, the green certificate unit price and the second green certificate unit price sold.
[0012] Optionally, when the energy processing model is an energy management scheduling model, the photovoltaic output data and the load data are constrained by a first power balance constraint condition, and the first power balance constraint condition includes:
[0013]
[0014] in, represents the net load at time t in the sth scenario; represents the original load at time t in the sth scenario; and They represent the charging and discharging power of the energy storage system at time t in the sth scenario respectively; and They represent the upward and downward power of the transferable load at time t in the sth scenario respectively; and They represent the wind energy, thermal power generation and solar energy at time t in the sth scenario respectively; represents the TGC requirement in the sth scenario; δ RPSRepresents the minimum amount of electricity that electricity users must consume from renewable energy as a percentage of their total load under the Renewable Portfolio Standard (RPS).
[0015] Optionally, the method further comprises:
[0016] Identify non-dispatchable loads, deferrable loads, multi-modal loads, and flexible loads on equipment in the target area;
[0017] The sum of the non-dispatchable load, the deferrable load, the multi-mode load and the flexible load is taken as the original load.
[0018] Optionally, when the energy processing model is a market clearing model, the operating cost of the power system is the sum of the electricity cost of conventional generators, the backup capacity cost and the probability of the target scenario occurring.
[0019] Optionally, when the energy processing model is a market clearing model, the photovoltaic output data and the load data are constrained by a second power balance constraint and / or a line flow constraint.
[0020] Optionally, the condition of the second power balance constraint includes:
[0021]
[0022] in, and They represent the power of conventional generator i and renewable energy source i at time t in the sth scenario respectively; Indicates the reserve power of conventional generators; represents the net load at time t in the sth scenario; and They represent the net load demand and the portion of electricity reserved by renewable energy sources, respectively.
[0023] Optionally, the line power flow constraint condition includes:
[0024]
[0025] Among them, m k-i represents the generation transfer allocation factor of line k; and They represent the power of conventional generator i, the power of renewable energy i and the reserve power of conventional generator i at time t in the sth scenario respectively; represents the net load at time t in the sth scenario; represents the transmission capacity of line k.
[0026] According to a second aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, on which computer program instructions are stored. When the program instructions are executed by a processor, the steps of the data processing method provided in the first aspect of the present disclosure are implemented.
[0027] The technical solution provided by the embodiments of the present disclosure may have the following beneficial effects:
[0028] The factors that affect the minimum operating cost, such as electricity price data, photovoltaic output data and load data, can be used as training samples to train the energy processing model. During the training process, when the operating cost output by the energy processing model based on the training samples in the next iteration is less than the operating cost output by the training samples in the previous iteration, and the difference between the two is less than the first preset difference, it means that the energy processing model can predict the minimum operating cost of the power system, and the energy processing model of this iteration can be used as the trained energy processing model.
[0029] Based on the trained energy processing model, after receiving the electricity price data, photovoltaic output data and load data of the actual operation of the power system, it can accurately predict the minimum operating cost of the power system while ensuring the operation of the power system.
[0030] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS
[0031] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.
[0032] Figure 1 The invention is a flowchart of a data processing method according to an exemplary embodiment.
[0033] Figure 2 It is a logical schematic diagram of a data processing method according to an exemplary embodiment.
[0034] Figure 3 It is a block diagram of a data processing device according to an exemplary embodiment.
[0035] Figure 4 It is a block diagram of a data processing device according to an exemplary embodiment. DETAILED DESCRIPTION
[0036] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.
[0037] It should be noted that all actions of acquiring signals, information or data in this application are carried out in compliance with the relevant data protection laws and policies of the country where they are located, and with the authorization given by the owner of the corresponding device.
[0038] Figure 1 is a flow chart of a data processing method according to an exemplary embodiment. Figure 1 As shown, the data processing method is used in a terminal and includes the following steps.
[0039] In step S11 , the energy processing model is trained using electricity price data, photovoltaic output data and load data as training samples.
[0040] Among them, electricity price data is the price of using electricity per unit time; photovoltaic output data is the power output of the photovoltaic power station; and load data is electricity consumption data.
[0041] In addition to electricity price data, photovoltaic output data and load data, the present invention can also obtain each group of transferable load data, solar energy equipment, thermal power units, wind turbines and energy storage capacity data in each local area power system, and use the K-means clustering algorithm to iteratively divide each group of the above data into K categories to obtain K power system operation scenarios. Each power system operation scenario corresponds to a group of data, and the data between each power system operation scenario are similar.
[0042] Among them, transferable load data refers to loads that are highly controllable. For example, transferable load data can be used to plan production in some production work, and production can be carried out when the electricity price is low. These transferable load data can be used for fire protection electricity, etc.
[0043] Energy storage capacity data refers to the relevant data of energy storage systems that can participate in scheduling within the local power system. The energy storage system includes battery energy storage, etc., which can store electricity when there is excess electricity in the local power system and release excess electricity during peak power consumption periods.
[0044] In step S12, during the iterative process of training the energy processing model, when the operating cost of the next iteration output of the energy processing model is less than the operating cost of the previous iteration output of the energy processing model, and the difference between the operating cost of the previous iteration output and the operating cost of the next iteration output is less than a first preset difference, the learnable parameters of the next iteration of the energy processing model are used as the target learnable parameters of the energy processing model to obtain a trained energy processing model, and the trained energy processing model is used to predict the minimum operating cost of the power system.
[0045] Among them, the energy processing model is used to predict the minimum operating cost of the power system, so as to save the operating cost of the power system while ensuring the smooth operation of each unit in the power system.
[0046] When the minimum operating cost output by the energy processing model in the last iteration is less than the minimum operating cost output by the energy processing model in the next iteration, and the difference between the two is less than the first preset difference, it means that the minimum operating cost output by the energy processing model in the next iteration can no longer continue to decrease. At this time, the operating cost output by the energy processing model is the minimum operating cost of the power system operation.
[0047] Among them, in each iteration process, the training samples input into the energy processing model are different, so as to ensure that after each iteration of the energy processing model, the output operating cost can be gradually reduced.
[0048] Among them, the learnable parameters include the locational marginal price (LMP) and the reserved capacity cost (ARC).
[0049] The node marginal price is composed of the Lagrange multipliers corresponding to the two constraints of the generator node power balance constraint and the line flow constraint. The calculation formula of the node marginal price is:
[0050]
[0051] in, is the node marginal electricity price; t,s The node marginal electricity price can be expressed by the Lagrange multiplier to achieve; is the Lagrange multiplier corresponding to the upper limit of the power transmitted by the transmission line; is the Lagrange multiplier corresponding to the lower limit of the transmission line power; G k-i For the transfer factor.
[0052] The calculation formula for spare capacity cost is:
[0053]
[0054] in, is the spare capacity cost; It indicates that the cost of reserve capacity can be realized by Lagrange multipliers; is the net load at time t under the Sth power system operation scenario; is the reserve power of conventional generator i at time t in the sth scenario.
[0055] Among them, in addition to the case where the operating cost of each output of the energy processing model cannot be further reduced, the learnable parameters obtained by the energy processing model each time, that is, the node marginal electricity price and the backup capacity cost cannot be further reduced, the learnable parameters obtained in this iteration can be used as the target learnable parameters to obtain the final energy processing model.
[0056] For example, see Figure 2 As shown, when the running cost of the next iteration output of the energy processing model is less than the running cost of the previous iteration output, and the difference between the running cost of the next iteration output and the running cost of the previous iteration output is less than the first preset difference; and the learnable parameters of the energy processing model in the next iteration process are less than the learnable parameters of the energy processing model in the previous iteration process, and the difference between the learnable parameters of the two iterations is less than the second preset difference, the learnable parameters in the next iteration process are used as the target learnable parameters to obtain the trained energy processing model. The first preset difference and the second preset difference in this process can be collectively referred to as convergence conditions.
[0057] The first preset difference value can be 0 or 10. -4 The present disclosure does not limit this.
[0058] Through the above technical solution, the factors such as electricity price data, photovoltaic output data and load data that affect the minimum operating cost can be used as training samples to train the energy processing model. During the training process, when the operating cost output by the energy processing model based on the training sample in the next iteration is less than the operating cost output by the training sample in the previous iteration, and the difference between the two is less than the first preset difference, it means that the energy processing model can predict the minimum operating cost of the power system, and the energy processing model of this iteration can be used as the trained energy processing model. Based on the trained energy processing model, after receiving the electricity price data, photovoltaic output data and load data of the actual operation of the power system, it can accurately predict the minimum operating cost of the power system under the premise of ensuring the operation of the power system.
[0059] In a possible implementation, the energy processing model includes an energy management scheduling model. Therefore, when the energy processing model includes the energy management scheduling model, the present disclosure further includes at least one of the following solutions:
[0060] Solution 1: When the energy processing model is an energy management and dispatching model, the operating cost of the power system is the sum of the power cost of the thermal power units, the green certificate cost, the transaction cost of the public power grid, the allocated reserve capacity cost and the load transfer cost. The following calculation formula (3) expresses the objective function of the energy management and dispatching model:
[0061]
[0062] in, is the electricity cost of thermal power units; for the cost of green certificates; transaction costs for the utility grid; The cost of the allocated spare capacity; The x parameter in the above five costs is the number provided by the local power system.
[0063] In some embodiments, when the actual green electricity consumption is less than the preset green electricity amount, the green certificate cost is obtained based on the preset green electricity amount, the green electricity consumption, the unit price of the first green certificate purchased, and the transaction penalty coefficient; when the green electricity consumption is greater than or equal to the preset green electricity amount, the green certificate cost is obtained based on the green electricity consumption, the green certificate unit price and the unit price of the second green certificate sold.
[0064] For example, the formula for green certificate cost is as follows:
[0065]
[0066] in, Indicates the first green certificate unit price of the green certificate purchased; C p is the transaction penalty coefficient; P res P is the preset green power; w is the green electricity consumption; The second green certificate unit price for the green certificates sold.
[0067] It can be seen from formula (4) that when the actual green electricity consumption is less than the preset green electricity consumption, it means that the green electricity consumption of the entire local power system does not meet the preset green electricity consumption. At this time, the difference between the green electricity consumption and the preset green electricity consumption can be divided by the preset value (for example, 1000), and then multiplied by the first green certificate unit price to obtain the first target value; then the difference between the preset green electricity consumption and the green electricity consumption is multiplied by the transaction penalty coefficient to obtain the second target value; finally, the first target value is subtracted from the second target value to obtain the green certificate cost. It can be understood that the green certificate cost obtained when the actual green electricity consumption is less than the preset green electricity consumption is the cost required to purchase the green certificate.
[0068] It can be seen from formula (4) that when the actual green electricity consumption is greater than or equal to the preset green electricity consumption, it means that the green electricity consumption of the entire local power system has met the preset green electricity consumption. At this time, the difference between the green electricity consumption and the preset green electricity consumption can be divided by the preset value (for example, 1000), and then multiplied by the second green certificate cost to obtain the green certificate cost. It can be understood that the green certificate cost obtained when the actual green electricity consumption is greater than or equal to the preset green electricity consumption is the income from the green certificate sold.
[0069] Optionally, the actual green electricity consumption can be increased by increasing the green electricity ratio so that the green electricity consumption reaches the preset green electricity amount, and then when the green electricity consumption reaches the preset green electricity amount, the green certificate cost caused by the additional purchase of green certificates can be reduced. The green electricity ratio refers to the proportion of green electricity such as wind power generation, hydropower generation, and solar power generation in the total power generation process of the local power system.
[0070] Solution 2: When the energy processing model is an energy management scheduling model, the photovoltaic output data and the load data are constrained by a first power balance constraint condition, and the first power balance constraint condition includes:
[0071]
[0072] in, represents the net load at time t in the sth scenario; represents the original load at time t in the sth scenario; and They represent the charging and discharging power of the energy storage system at time t in the sth scenario respectively; and They represent the upward and downward power of the transferable load at time t in the sth scenario respectively; and They represent the wind energy, thermal power generation and solar energy at time t in the sth scenario respectively; represents the TGC requirement in the sth scenario; δ RPS It indicates the percentage of the minimum amount of electricity consumed by power users from renewable energy as a percentage of the total load required by the RPS (Renewable Portfolio Standard). The scenario S in the above parameters refers to the operation scenario of the power system.
[0073] The first equation in formula (5) indicates that the net load is equal to the sum of the original load, the difference between the charging and discharging power of the energy storage system, and the difference between the up and down power of the transferable load, minus the sum of the power generation functions of wind power, thermal power units, and solar energy; the second equation indicates the TGC (Tradable Green Certificate) demand of the power system in the Sth scenario, which indicates the minimum amount of green electricity that the local power system should have access to the grid.
[0074] Among them, the electricity price data is a known constant and does not need to be constrained by the first power constraint. The electricity price data can affect the transaction cost of the public power grid in the objective function of the energy management scheduling model.
[0075] In some embodiments, an unschedulable load, a deferrable load, a multi-mode load and a flexible load of equipment in a target area are determined; and the sum of the unschedulable load, the deferrable load, the multi-mode load and the flexible load is taken as the original load.
[0076] There are multiple devices in the target area, and each device is of a different type, so the load type required by each device is also different. For example, the load required by the equipment in the hospital is a non-schedulable load, and the load required by the equipment in the residential building is a flexible load, etc.
[0077] Optionally, the non-dispatchable load refers to the load that cannot be dispatched by the local power system. For example, the equipment in the hospital needs to use electrical energy, so the load required by the equipment in the hospital is the non-dispatchable load. The non-dispatchable load is usually regarded as a constant.
[0078] Optionally, a deferrable load is a load that is assigned to complete a task in a specified time period, but it is not specified which target time period in the specified time period is to complete the task, so this part of the deferrable load can be completed in a delayed manner. For example, the deferrable load is specified to be used on A day, month, year A. Since the electricity cost is higher in the morning of A, this part of the deferrable load can be delayed to the afternoon of A.
[0079] Optionally, the deferrable load has the following constraints:
[0080]
[0081] Among them, Pj () is the power consumption of the delayable load j in the time interval τ, and the time interval τ is the scheduling time interval of the delayable load, for example, the delayable load is transferred every 8 hours; W j is the duration of the load curve of the delayable load j; P j,t is the deferrable load at time t.
[0082] Optionally, the multi-mode load refers to the load of the device in different modes. For example, an air conditioner has a sleep mode and a cooling mode. In the sleep mode, the air conditioner requires one load, while in the cooling mode, the air conditioner requires a greater load.
[0083] Optionally, the multi-mode load has the following constraints:
[0084]
[0085] Among them, P k,m,t is the load of the multi-mode load K in mode m in time interval t, and time interval t is the time interval for switching between multi-mode loads, for example, the air conditioner switches from sleep mode to cooling mode every 1 hour; It is a ratio. Different modes correspond to different ratios.
[0086] It can be seen from formula (7) that the load of the device in different modes is added together to obtain the overall multi-mode load of the device. For example, the air conditioner has a cooling mode and a sleep mode. The load required by the air conditioner in the cooling mode and the load required by the air conditioner in the sleep mode are added together to obtain the multi-mode load.
[0087] It can be understood that the load proposed in any embodiment of the present disclosure is actually the power consumption required by the device, which can also be called the power required by the device.
[0088] Optionally, flexible loads are flexible loads that residential users can adjust flexibly in each time period, and these flexible loads can be changed by users' adjustments. Flexible loads are subject to conditions such as minimum continuous load duration constraints, maximum continuous shutdown duration constraints, all-day load constraints, load change rate constraints, and maximum and minimum load constraints.
[0089] Among them, the minimum continuous load duration constraint means that the actual working time of the equipment must be greater than or equal to the set minimum continuous load duration. For example, for user electricity, if the power is cut off during the specified power consumption time, it may cause power waste or reduce the user's quality of life. Therefore, this part of the load needs to be constrained. The specific constraints are as follows:
[0090]
[0091] In formula (8), ui,t-1 The working status of the device at the last moment, for example, u i,t-1 When it is 1, it means the device is turned on. i,t-1 When it is 0, it means the device is turned off; i,t The working status of the device at the current moment; The actual working hours accumulated for the equipment; It is the preset minimum continuous load time.
[0092] From formula (8), it can be seen that when the device is turned on at the previous moment and turned off at the current moment, u i,t-1 -u i,t =1. At this time, if the actual working time accumulated by the equipment is less than the preset minimum continuous load time, the actual working time accumulated by the equipment does not meet the demand; if the actual working time accumulated by the equipment is greater than or equal to the preset minimum continuous load time, the actual working time accumulated by the equipment meets the demand. Formula (8) can limit the actual working time accumulated by the equipment to be greater than the minimum continuous load time, thereby ensuring that these devices can be powered for a longer time, thereby improving the quality of life of users.
[0093] The maximum continuous downtime limit refers to the limit on the continuous downtime of the device, and the continuous downtime of the device needs to be less than or equal to the maximum continuous downtime. For example, for some appliances such as air conditioners and electric heaters that are used to maintain indoor temperature, if the load of such equipment stops working for a period of time, the indoor temperature will change due to the outdoor temperature, which will reduce the quality of life of users. Therefore, it is necessary to limit the maximum continuous downtime of these devices, as follows:
[0094]
[0095] In formula (9), The maximum continuous downtime duration of the preset load; u i,τ It is the working status of the device at the current moment.
[0096] From formula (9), it can be seen that in the working state u i,τ When it is 0, it means that the device stops working. If the cumulative stop working time when the device working state is 0 is less than or equal to the preset maximum continuous stop time, it means that the device stop working time meets the requirements; if the cumulative stop working time when the device working state is 0 is greater than the preset maximum continuous stop time, it means that the device stop working time does not meet the requirements. Through formula (9), the stop working time of the device can be limited to the preset maximum continuous stop time, so that the stop working time of this part of the equipment is shorter, thereby meeting the power demand of the equipment to a greater extent.
[0097] Among them, the all-day load limit means that for some devices, they can be powered for a longer time. For example, for electric scooters, users may need to charge the batteries of electric scooters. If the power is cut off for these devices, the electric scooters will undoubtedly fail to be fully charged. Based on this, the load of these devices can be limited, as follows:
[0098]
[0099] In formula (10), P i,t is the load actually required by the equipment; E i The preset load.
[0100] It can be seen from formula (10) that the cumulative load required by the device at different times needs to be greater than the preset load to meet the requirements. Through formula (10), the actual load required by the device can be limited to above the preset load to meet the long-term charging needs of this part of the device.
[0101] Among them, the load change rate limit is similar to the generator climbing constraint. For some household electrical appliances, there are certain requirements for the load change rate. It is necessary to ensure that the load change rate of these devices is kept within a certain range to avoid burnout of these devices due to excessive load change rate. The details are as follows:
[0102]
[0103]
[0104] In formula (11), is the maximum load change rate of increase; is the maximum load change rate of reduction; P i,t is the load of the equipment at the current moment; P i,t-1 is the load of the equipment at the last moment.
[0105] It can be seen from formula (11) that the load change rate of the equipment that is actually increased and the load change rate of the equipment that is actually reduced can be limited within the maximum load change rate to avoid the load change rate of this part of the equipment being too large and causing ablation of this part of the equipment.
[0106] Among them, most equipment has maximum and minimum load limits, as follows:
[0107]
[0108] In formula (12), is the minimum load of the equipment at time t; is the maximum load of the equipment at time t; P i,t The actual load generated by the equipment.
[0109] It can be seen from formula (12) that the actual load of the equipment can be limited to the range of maximum load and minimum load.
[0110] In Scheme 1 and Scheme 2, the first power balance constraint constrains the photovoltaic output data and load data in the objective function of the energy management and dispatching model. Based on the photovoltaic output data and load data, the energy management and dispatching model can predict the sum of the power cost of the thermal power unit, the green certificate cost, the allocated reserve capacity cost and the cost of load transfer; based on the electricity price data, the energy management and dispatching model can predict the transaction cost of the public power grid. Therefore, the energy management and dispatching model can predict the minimum operating cost of the entire power system based on the electricity price data, photovoltaic output data and load data, with the goal of minimizing the objective function (3).
[0111] Through the above technical scheme, when the energy processing model is an energy management and scheduling model, after the photovoltaic output data and the load data are constrained by the first power balance constraint, the energy management and scheduling model can output the sum of the electricity cost of the thermal power unit, the green certificate cost, the transaction cost of the public power grid, the shared reserve capacity cost and the load transfer cost. When the sum reaches the minimum, the energy management and scheduling model can output the minimum operating cost of the power system.
[0112] Moreover, the original load proposed in the present disclosure is not a fixed original load, but is flexibly configured according to the different types of equipment in the target area. The final original load is closer to the actual situation of the target area, so the first power balance constraint obtained based on the actual original load is also more accurate. Furthermore, the photovoltaic output data and load data obtained based on the first power balance constraint are also more in line with reality.
[0113] In addition, the green certificate cost is based on the actual green electricity consumption, so by minimizing the green certificate cost in formula (3), a more reasonable green electricity consumption can be obtained, thereby providing guidance on green electricity consumption for the local power system. The local power system can generate a smaller green certificate cost by using this specific green electricity consumption.
[0114] In a possible implementation, the energy processing model includes a market clearing model. Therefore, in the case where the energy processing model includes a market clearing model, the present disclosure further includes at least one of the following solutions:
[0115] Solution 3: When the energy processing model is a market clearing model, the operating cost of the power system is the sum of the power cost of conventional generators, the backup capacity cost, and the probability of the target scenario occurring. The following calculation formula (13) expresses the objective function of the market clearing model:
[0116]
[0117] in, represents the conventional generator cost; represents the spare capacity cost; γ s is the probability of scene s appearing.
[0118] Solution 4: When the energy processing model is a market clearing model, the photovoltaic output data and the load data are constrained by a second power balance constraint and / or a line flow constraint.
[0119] The conditions of the second power balance constraint include:
[0120]
[0121]
[0122] in, and They represent the power of conventional generator i and renewable energy source i at time t in the sth scenario respectively; Indicates the reserve power of conventional generators; represents the net load at time t in the sth scenario; and They represent the net load demand and the portion of electricity reserved by renewable energy sources, respectively.
[0123] The conditions of the line flow constraint include:
[0124]
[0125] Among them, m k-i represents the generation transfer allocation factor of line k; and They represent the power of conventional generator i, the power of renewable energy i and the reserve power of conventional generator i at time t in the sth scenario respectively; represents the net load at time t in the sth scenario; represents the transmission capacity of line k; is the Lagrange multiplier corresponding to the upper limit of the power transmitted by the transmission line; is the Lagrange multiplier corresponding to the lower limit of the power transmitted by the transmission line.
[0126] Among them, the electricity price data is a known constant and does not need to be constrained by the second power constraint and the line flow constraint. The electricity price data can affect the cost of conventional generators in the public power grid in the objective function of the market clearing model.
[0127] In Scheme 3 and Scheme 4, the second power balance constraint constrains the photovoltaic output data and load data in the objective function of the market clearing model. The market clearing model can predict the sum of the cost of conventional generators and the cost of backup capacity based on the photovoltaic output data and load data. The line flow constraint constrains the upper and lower limits of the generator power, the upper and lower limits of the power source, and the ramp constraint, etc., to constrain the power transmitted on the transmission line. Therefore, the energy management scheduling model can predict the minimum operating cost of the entire power system based on electricity price data, photovoltaic output data, and load data, with the goal of minimizing the objective function (13).
[0128] Through the above technical solution, when the energy processing model is a market-clearing model, after the photovoltaic output data and the load data are constrained by the second power balance constraint, the market-clearing model can output the sum of the conventional generator cost and the backup capacity cost. When the sum reaches the minimum, it is considered that the market-clearing model can output the minimum operating cost of the power system.
[0129] Figure 3 is a block diagram of a data processing device according to an exemplary embodiment. Figure 3 The data processing device 300 includes a training module 310 and an iteration module 320 .
[0130] The training module 310 is configured to train the energy processing model using electricity price data, photovoltaic output data and load data as training samples;
[0131] The iteration module 320 is configured to, during the iterative process of training the energy processing model, use the learnable parameters of the next iteration of the energy processing model as the target learnable parameters of the energy processing model when the operating cost of the next iteration output of the energy processing model is less than the operating cost of the previous iteration output of the energy processing model, and the difference between the operating cost of the previous iteration output and the operating cost of the next iteration output is less than a first preset difference, to obtain a trained energy processing model, and the trained energy processing model is used to predict the minimum operating cost of the power system.
[0132] Optionally, when the energy processing model is an energy management scheduling model, the operating cost of the power system is the sum of the power cost of the thermal power units, the green certificate cost, the transaction cost of the public power grid, the shared reserve capacity cost and the load transfer cost.
[0133] Optionally, the data processing device 300 further includes:
[0134] A first green certificate cost determination module is configured to obtain the green certificate cost according to the preset green power quantity, the green power consumption, the unit price of the purchased first green certificate and the transaction penalty coefficient when the actual green power consumption is less than the preset green power quantity;
[0135] The second green certificate cost determination module is configured to obtain the green certificate cost according to the green electricity consumption, the green certificate unit price and the second green certificate unit price sold when the green electricity consumption is greater than or equal to the preset green electricity consumption.
[0136] Optionally, when the energy processing model is an energy management scheduling model, the photovoltaic output data and the load data are constrained by a first power balance constraint condition, and the first power balance constraint condition includes:
[0137]
[0138] in, represents the net load at time t in the sth scenario; represents the original load at time t in the sth scenario; and They represent the charging and discharging power of the energy storage system at time t in the sth scenario respectively; and They represent the upward and downward power of the transferable load at time t in the sth scenario respectively; and They represent the wind energy, thermal power generation and solar energy at time t in the sth scenario respectively; represents the TGC requirement in the sth scenario; δ RPS Represents the minimum amount of electricity that electricity users must consume from renewable energy as a percentage of their total load under the Renewable Portfolio Standard (RPS).
[0139] Optionally, the data processing device 300 further includes:
[0140] A multi-type load determination module is configured to determine non-dispatchable loads, deferrable loads, multi-mode loads, and flexible loads of equipment in a target area;
[0141] The original load calculation module is configured to take the sum of the unschedulable load, the deferrable load, the multi-mode load and the flexible load as the original load.
[0142] Optionally, when the energy processing model is a market clearing model, the operating cost of the power system is the sum of the electricity cost of conventional generators, the backup capacity cost and the probability of the target scenario occurring.
[0143] Optionally, when the energy processing model is a market clearing model, the photovoltaic output data and the load data are constrained by a second power balance constraint and / or a line flow constraint.
[0144] Optionally, the condition of the second power balance constraint includes:
[0145]
[0146] in, and They represent the power of conventional generator i and renewable energy source i at time t in the sth scenario respectively; Indicates the reserve power of conventional generators; represents the net load at time t in the sth scenario; and They represent the net load demand and the portion of electricity reserved by renewable energy sources, respectively.
[0147] Optionally, the line power flow constraint condition includes:
[0148]
[0149] Among them, m k-i represents the generation transfer allocation factor of line k; and They represent the power of conventional generator i, the power of renewable energy i and the reserve power of conventional generator i at time t in the sth scenario respectively; represents the net load at time t in the sth scenario; represents the transmission capacity of line k.
[0150] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.
[0151] The present disclosure also provides a computer-readable storage medium on which computer program instructions are stored. When the program instructions are executed by a processor, the steps of the data processing method provided by the present disclosure are implemented.
[0152] In addition to being an independent electronic device, the above-mentioned device can also be a part of an independent electronic device. For example, in one embodiment, the device can be an integrated circuit (IC) or a chip, wherein the integrated circuit can be an IC or a collection of multiple ICs; the chip can include but is not limited to the following types: GPU (Graphics Processing Unit), CPU (Central Processing Unit), FPGA (Field Programmable Gate Array), DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), SOC (System on Chip, SoC), etc. The above-mentioned integrated circuit or chip can be used to execute executable instructions (or codes) to implement the above-mentioned data processing method. The executable instructions can be stored in the integrated circuit or chip, or can be obtained from other devices or equipment, such as the integrated circuit or chip including a processor, a memory, and an interface for communicating with other devices. The executable instruction can be stored in the memory, and when the executable instruction is executed by the processor, the above-mentioned data processing method is implemented; alternatively, the integrated circuit or chip can receive the executable instruction through the interface and transmit it to the processor for execution, so as to implement the above-mentioned data processing method.
[0153] In another exemplary embodiment, a computer program product is also provided. The computer program product includes a computer program executable by a programmable device, and the computer program has a code portion for executing the above data processing method when executed by the programmable device.
[0154] Figure 4 1 is a block diagram of a device 1900 for data processing according to an exemplary embodiment. For example, the device 1900 may be provided as a server. Figure 4 , the apparatus 1900 includes a processing component 1922, which further includes one or more processors, and a memory resource represented by a memory 1932 for storing instructions, such as an application, that can be executed by the processing component 1922. The application stored in the memory 1932 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 1922 is configured to execute instructions to perform the above method.
[0155] The device 1900 may also include a power supply component 1926 configured to perform power management of the device 1900, a wired or wireless network interface 1950 configured to connect the device 1900 to a network, and an input / output interface 1958. The device 1900 may operate based on an operating system stored in the memory 1932.
[0156] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the present disclosure. This application is intended to cover any variations, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure. The specification and examples are to be considered as exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.
[0157] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims
1. A data processing method, characterized in that: include: The energy processing model is trained using electricity price data, photovoltaic output data, and load data as training samples; During the iterative process of training the energy processing model, when the operating cost of the next iteration output of the energy processing model is less than the operating cost of the previous iteration output of the energy processing model, and the difference between the operating cost of the previous iteration output and the operating cost of the next iteration output is less than a first preset difference, the learnable parameters of the next iteration of the energy processing model are used as the target learnable parameters of the energy processing model to obtain a trained energy processing model, and the trained energy processing model is used to predict the minimum operating cost of the power system.
2. The method according to claim 1, characterized in that When the energy processing model is an energy management and scheduling model, the operating cost of the power system is the sum of the power cost of the thermal power units, the green certificate cost, the transaction cost of the public power grid, the shared reserve capacity cost and the load transfer cost.
3. The method according to claim 2, characterized in that The method further comprises: When the actual amount of green electricity consumed is less than the preset amount of green electricity, the green electricity consumption, the unit price of the first green certificate purchased, and the transaction penalty coefficient, the green certificate cost is obtained; When the green electricity consumption is greater than or equal to the preset green electricity amount, the green certificate cost is obtained according to the green electricity consumption, the green certificate unit price and the second green certificate unit price sold.
4. The method according to claim 2, characterized in that: In the case where the energy processing model is an energy management scheduling model, the photovoltaic output data and the load data are constrained by a first power balance constraint condition, and the conditions of the first power balance constraint include: in, represents the net load at time t in the sth scenario; represents the original load at time t in the sth scenario; and They represent the charging and discharging power of the energy storage system at time t in the sth scenario respectively; and They represent the upward and downward power of the transferable load at time t in the sth scenario respectively; and They represent the wind energy, thermal power generation and solar energy at time t in the sth scenario respectively; represents the TGC requirement in the sth scenario; δ RPS Represents the minimum amount of electricity that electricity users must consume from renewable energy as a percentage of their total load under the Renewable Portfolio Standard (RPS).
5. The method according to claim 4, characterized in that The method further comprises: Identify non-dispatchable loads, deferrable loads, multi-modal loads, and flexible loads on equipment in the target area; The sum of the non-dispatchable load, the deferrable load, the multi-mode load and the flexible load is taken as the original load.
6. The method according to claim 1, characterized in that In the case where the energy processing model is a market clearing model, the operating cost of the power system is the sum of the power cost of conventional generators, the backup capacity cost, and the probability of the target scenario occurring.
7. The method according to claim 6, characterized in that In the case where the energy processing model is a market clearing model, the photovoltaic output data and the load data are constrained by a second power balance constraint and / or a line flow constraint.
8. The method according to claim 7, characterized in that The conditions of the second power balance constraint include: in, and They represent the power of conventional generator i and renewable energy source i at time t in the sth scenario respectively; Indicates the reserve power of conventional generators; represents the net load at time t in the sth scenario; and They represent the net load demand and the portion of electricity reserved by renewable energy sources, respectively.
9. The method according to claim 7, characterized in that: The line flow constraint conditions include: Among them, m k-i represents the generation transfer allocation factor of line k; and They represent the power of conventional generator i, the power of renewable energy i and the reserve power of conventional generator i at time t in the sth scenario respectively; represents the net load at time t in the sth scenario; represents the transmission capacity of line k.
10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the program instructions are executed by a processor, the steps of the method described in any one of claims 1 to 9 are implemented.