Intelligent power grid dispatching method and device with independent energy storage and new energy participation

Abstract

The invention provides an intelligent power grid dispatching method and device with independent energy storage and new energy participation, and belongs to the field of power grids. The method comprises the following steps: predicting the generation power of new energy to obtain the predicted generation power of the new energy; determining a segmented quotation model of the new energy based on the new energy prediction generation power; the segmented quotation model of the new energy constrains a first quotation coefficient of the new energy, so that the quotation of the new energy is the lowest when the ratio of the actual power generation power to the predicted power generation power of the new energy is a preset ratio; based on the segmented quotation model of the new energy, determining an SCUC clearing calculation model, and solving the SCUC clearing calculation model to obtain a combination of a power generation unit and an energy storage unit; based on the combination of the power generation unit and the energy storage unit and a predetermined SCED model, solving to obtain a power generation plan and a node electricity price; and performing power grid scheduling based on the power generation plan and the node electricity price. According to the invention, the new energy power abandoning rate can be effectively reduced.

Description

Technical Field

[0001] The present invention relates to the field of power grid technology, and in particular to a smart grid dispatching method and device involving independent energy storage and new energy. Background Art

[0002] Renewable energy generation generally has the problem of unstable power generation. In order to solve this problem, energy storage units can participate in the power market. Energy storage units can achieve efficient utilization, conversion and storage of electric energy. They can store energy during the peak period of renewable energy generation and supplement its power generation during the valley period, achieving the effect of "shifting peaks and filling valleys". Energy storage units help improve the flexibility of power system operation, promote the consumption of renewable energy, and reduce the phenomenon of wind and solar power abandonment.

[0003] In the relevant technologies, the mechanism for new energy to participate in the electricity market clearing still continues the traditional thermal power market model, and does not make good use of the various characteristics of new energy. In the new power system under construction, new energy will occupy the dominant position in electricity and electricity volume. The traditional electricity market model will not be able to meet the competition in the new market, which may easily lead to a higher rate of new energy power abandonment. Summary of the invention

[0004] The embodiments of the present invention provide a method and device for smart grid dispatching involving independent energy storage and new energy, so as to solve the problem that the existing methods easily lead to a high rate of new energy abandonment.

[0005] In a first aspect, an embodiment of the present invention provides a smart grid dispatching method involving independent energy storage and new energy, including:

[0006] Predict the power generation of new energy to obtain the predicted power generation of new energy;

[0007] Based on the predicted power generation of new energy, a segmented bidding model for new energy is determined; the segmented bidding model for new energy constrains the first bidding coefficient of new energy, so that when the ratio of the actual power generation of new energy to the corresponding predicted power generation is a preset ratio, the bid of new energy is the lowest;

[0008] Based on the segmented bidding model of new energy, the SCUC clearing calculation model is determined and solved to obtain the combination of power generation and energy storage units;

[0009] Based on the combination of power generation and energy storage units and the predetermined SCED model, the power generation plan and node electricity price are solved;

[0010] Grid dispatching is carried out based on power generation plans and node electricity prices.

[0011] In a possible implementation, the constraint conditions for constraining the first bidding coefficient of the new energy by the segmented bidding model of the new energy include:

[0012]

[0013] Among them, a i is the first quotation coefficient of the i-th quotation segment; P new,i is the actual power generation of renewable energy in the i-th quotation period; The predicted power generation of renewable energy in the i-th quotation period; the preset ratio is 0.9.

[0014] In a possible implementation, the segmented quotation model for new energy also includes:

[0015] C new,i (P new,i )=a i P new,i +b i

[0016] y i P min,i ≤P new,i ≤y i P max,i

[0017]

[0018] Among them, b i is the second quotation coefficient of the i-th quotation segment; C new,i (P new,i ) The price of new energy in the i-th quotation period; y i P is a 0-1 variable indicating whether the new energy output reported in the i-th quotation period is called; min,i is the minimum power generation of renewable energy in the i-th quotation period; P max,i is the maximum power generation of renewable energy in the i-th quotation period; n is the number of quotation periods; P new is the total power generation of renewable energy.

[0019] In a possible implementation, based on the segmented quotation model of new energy, the SCUC clearing calculation model is determined, including:

[0020] Calculate the environmental costs of thermal power plants;

[0021] Based on the segmented bidding model of new energy and the environmental cost of thermal power units, the SCUC clearing calculation model is determined with the goal of minimizing cost.

[0022] In one possible implementation, the SCUC clearing calculation model includes a first objective function with the lowest cost as the goal and a first constraint condition; the first constraint condition includes a thermal power unit constraint, a line flow constraint, a node electricity price calculation constraint, and an energy storage device constraint.

[0023] In a possible implementation, the first objective function includes:

[0024]

[0025] Among them, C e,k is the environmental cost of the kth thermal power unit; C k,t (P k,t ) is the operating cost of the kth thermal power unit in the tth period; is the electricity price reported by the kth thermal power unit in the ith quotation period of the tth period, P k,t,i The declared output of the kth thermal power unit in the ith quotation period during the tth period; is the startup cost of the kth thermal power unit in the tth period; μ k,t is a 0-1 variable indicating the start and stop of the kth thermal power unit in the tth period, is the startup cost of the kth thermal power unit; is the discharge price of the es-th energy storage unit in the t-th period; is the charging price of the es-th energy storage unit in the t-th period; is the discharge power of the es-th energy storage unit in the t-th period; is the charging power of the es-th energy storage unit in the t-th period; C new,lt is the cost of the lth renewable energy unit in the tth period; β is the power abandonment penalty coefficient; is the power abandonment of the lth renewable energy unit in the tth period; min f is the first objective function.

[0026] In one possible implementation,

[0027] in, is the predicted power generation of the lth new energy unit in the tth period; is the actual power generation of the lth new energy unit in the tth period.

[0028] In a possible implementation, the SCED model includes a second objective function and a second constraint;

[0029] The second objective function includes:

[0030]

[0031] Among them, C k,t (P k,t ) is the operating cost of the kth thermal power unit in the tth period; is the discharge price of the es-th energy storage unit in the t-th period; is the charging price of the es-th energy storage unit in the t-th period; is the discharge power of the es-th energy storage unit in the t-th period; is the charging power of the es-th energy storage unit in the t-th period; C new,lt is the cost of the lth renewable energy unit in the tth period; β is the power abandonment penalty coefficient; is the power abandonment of the lth new energy unit in the tth period; min E is the second objective function.

[0032] In a possible implementation, predicting the power generation of new energy to obtain the predicted power generation of new energy includes:

[0033] Based on the pre-trained new energy prediction deep learning model, the power generation of new energy is predicted to obtain the predicted power generation of new energy; the new energy prediction deep learning model includes convolution layer, pooling layer, bidirectional long short-term memory layer and fully connected layer.

[0034] In a second aspect, an embodiment of the present invention provides a smart grid dispatching device with independent energy storage and new energy participation, including:

[0035] A prediction module is used to predict the power generation of new energy and obtain the predicted power generation of new energy;

[0036] A new energy quotation model determination module is used to determine a segmented quotation model of new energy based on the predicted power generation of new energy; the segmented quotation model of new energy constrains the first quotation coefficient of new energy so that when the ratio of the actual power generation of new energy to the corresponding predicted power generation is a preset ratio, the quotation of new energy is the lowest;

[0037] The first solution module is used to determine the SCUC clearing calculation model based on the segmented quotation model of new energy, and solve the SCUC clearing calculation model to obtain the combination of power generation and energy storage units;

[0038] The second solution module is used to solve the power generation plan and node electricity price based on the combination of power generation and energy storage units and a predetermined SCED model;

[0039] The scheduling module is used to schedule the power grid based on the power generation plan and node electricity prices.

[0040] The embodiment of the present application provides a smart grid dispatching method and device involving independent energy storage and new energy. The method determines a segmented quotation model of new energy based on the predicted power generation of new energy, and the segmented quotation model of new energy constrains the first quotation coefficient of new energy, so that when the ratio of the actual power generation of new energy to the predicted power generation is a preset ratio, the quotation of new energy is the lowest. Then, based on the segmented quotation model of new energy, the SCUC clearing calculation model is determined, thereby taking into account the market clearing mechanism of new energy participating in the electricity market. The combination of power generation and energy storage units and the power generation plan obtained based on this solution can effectively reduce the power abandonment rate of new energy. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 It is a flow chart of an implementation of a smart grid dispatching method involving independent energy storage and new energy provided by an embodiment of the present invention;

[0043] Figure 2 It is a schematic diagram of training and predicting curves of a new energy prediction deep learning model provided by an embodiment of the present invention;

[0044] Figure 3 It is a schematic diagram of using a trained new energy prediction deep learning model to predict the power generation of new energy in the next 24 hours, provided by an embodiment of the present invention;

[0045] Figure 4 is a schematic diagram of a load curve of a system in a day in a calculation example provided by an embodiment of the present invention;

[0046] Figure 5 is a schematic diagram of network parameters of a system in a calculation example provided in an embodiment of the present invention;

[0047] Figure 6 It is a schematic diagram of the winning bid of a thermal power unit after clearance provided by an embodiment of the present invention;

[0048] Figure 7 It is a schematic diagram of the winning bid of the energy storage unit after clearance provided by one embodiment of the present invention;

[0049] Figure 8 It is a schematic diagram of the winning bid of the new energy unit after the clearance provided by an embodiment of the present invention;

[0050] Fig. 9is a schematic diagram of the power abandonment rate of new energy provided by an embodiment of the present invention;

[0051] Fig.10 It is a structural diagram of a smart grid dispatching device involving independent energy storage and new energy provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0053] In order to make the purpose, technical solutions and advantages of the present invention more clear, specific embodiments will be described below in conjunction with the accompanying drawings.

[0054] As mentioned above, only typical models and parameters can be used in transient simulation calculations of power grids, which makes it difficult to ensure the accuracy of the energy storage power station simulation model, thereby affecting the credibility of large-scale energy storage grid-connected simulation results. In addition, the data used for parameter identification of energy storage converters mainly comes from laboratory "all-digital" model simulation and field testing. Since the experimental conditions of "all-digital" model simulation are too ideal, the parameters identified are difficult to use in engineering practice; and field testing needs to be carried out under specific circumstances, and the cost of obtaining experimental data is high. For many types of energy storage converters, it is difficult to obtain experimental data through field tests one by one.

[0055] In order to solve the above problems, an embodiment of the present invention provides a smart grid scheduling method involving independent energy storage and new energy.

[0056] Figure 1 The following is a flowchart of a smart grid dispatching method for independent energy storage and new energy participation provided in an embodiment of the present invention. The smart grid dispatching method for independent energy storage and new energy participation is described in detail as follows:

[0057] Step 101, predicting the power generation of new energy to obtain the predicted power generation of new energy.

[0058] The embodiment of the present application can predict the power generation of new energy in the future period of time to obtain the predicted power generation of new energy. The predicted power generation of new energy can include the power generation of new energy in each period of time in the future period of time. The length of the future period of time and the length of a single period of time can be set according to actual needs, and no specific restrictions are made here. Among them, the power generation can also be called output.

[0059] The embodiments of the present application do not impose any specific restrictions on the specific means of implementing the prediction of the power generation power of new energy, and any achievable means may be used.

[0060] Step 102, based on the predicted power generation of new energy, determine the segmented bidding model of new energy; the segmented bidding model of new energy constrains the first bidding coefficient of new energy, so that when the ratio of the actual power generation of new energy to the corresponding predicted power generation is a preset ratio, the bid of new energy is the lowest.

[0061] The embodiment of the present application can construct a segmented quotation model for new energy sources based on the predicted power generation of new energy sources determined above. The segmented quotation model for new energy sources can quote low prices in the low power stage, slightly higher prices in the medium power stage, and continue to increase prices in the high power stage to prevent excessive electricity from being concentrated on new energy generation and causing system instability. In the segmented quotation model for new energy sources, when the ratio of the actual power generation of new energy to the predicted power generation of the corresponding quotation segment is a preset ratio, the quotation for new energy sources is the lowest, thereby increasing the revenue of new energy sources and reducing the power abandonment rate.

[0062] Step 103, based on the segmented bidding model of new energy, determine the SCUC (Security-Constrained Unit Commitment) clearing calculation model, and solve the SCUC clearing calculation model to obtain a combination of power generation and energy storage units.

[0063] The above combination of power generation and energy storage units refers to the combination of power generation units and energy storage units that won the bid. The power generation units may include new energy units and thermal power units.

[0064] The embodiment of the present application can determine the SCUC clearing calculation model based on the segmented bidding model of new energy, and the model can include relevant information of new energy units, energy storage units and thermal power units. Then, by solving the SCUC clearing calculation model, the winning combination of power generation units and energy storage units can be obtained.

[0065] A penalty mechanism for power abandonment of renewable energy is added to the SCUC clearing calculation model, and a penalty term is added to the corresponding first objective function to control the power abandonment phenomenon of renewable energy. This can effectively improve the problem of wind and solar power abandonment of renewable energy, while having good solution efficiency.

[0066] Step 104, based on the combination of power generation and energy storage units and a predetermined SCED (Security-Constrained Economic Dispatch) model, solve to obtain a power generation plan and a node electricity price.

[0067] Based on the combination of power generation and energy storage units and the pre-determined SCED model, this application can solve the specific power generation plan of the winning power generation unit and the node electricity price of each line node. The above power generation plan can also be called an output plan.

[0068] In the SCED model, a penalty mechanism for the abandonment of renewable energy can also be introduced. A penalty term can be added to the corresponding second objective function to control the abandonment of renewable energy. This can effectively improve the abandonment of wind and solar power and has good solution efficiency.

[0069] In some possible implementations, after the above step 104, the following may also be included:

[0070] Based on the above power generation plan and node electricity price, the profit situation of each party in the power market is calculated and output. The profit situation of each party can include the profit situation of new energy, the profit situation of energy storage and the profit situation of thermal power units.

[0071] Step 105: Perform grid dispatching based on the power generation plan and node electricity prices.

[0072] The embodiment of the present application can perform corresponding grid scheduling based on the above-determined power generation plan and node electricity prices to perform corresponding optimization.

[0073] The embodiment of the present application determines a segmented bidding model for new energy based on the predicted power generation of new energy, and the segmented bidding model for new energy constrains the first bidding coefficient of new energy, so that when the ratio of the actual power generation of new energy to the predicted power generation is a preset ratio, the bid of new energy is the lowest. Then, based on the segmented bidding model of new energy, the SCUC clearing calculation model is determined, thereby taking into account the market clearing mechanism of new energy participating in the electricity market. The combination of power generation and energy storage units and the power generation plan obtained based on this solution can effectively reduce the power abandonment rate of new energy.

[0074] In some embodiments, the above step 101 may include:

[0075] Based on the pre-trained new energy prediction deep learning model, the power generation of new energy is predicted to obtain the predicted power generation of new energy; the new energy prediction deep learning model includes convolution layer, pooling layer, bidirectional long short-term memory layer and fully connected layer.

[0076] In an embodiment of the present application, the new energy prediction deep learning model is a deep learning model that combines a convolutional neural network and a bidirectional long short-term memory network, and may include a convolutional layer, a pooling layer, a bidirectional long short-term memory layer, and a fully connected layer connected in sequence.

[0077] The training process of the deep learning model for new energy prediction can include:

[0078] Acquire historical power generation data of different groups of new energy units, and convert the historical power generation data into matrix data; wherein different groups of new energy units correspond to different historical power generation data, and the historical power generation data may include weather data of multiple historical periods and / or power generation data of corresponding new energy units, etc.;

[0079] Normalize the matrix data, scale it to the interval [0,1], set the time step, convert the normalized data into the data form of supervised learning, and divide the converted data set into training set and test set;

[0080] The pre-built deep learning model for new energy prediction is trained based on the training set, and the test set is used to evaluate the trained deep learning model for new energy prediction. The deep learning model for new energy prediction selects Nadam as the optimizer to achieve better convergence speed and search capability, and selects MSE as the loss function. During the training process, the learning rate scheduler is used to prevent falling into the local optimum, and the early stopping mechanism is added to prevent overfitting, and the number of training rounds and batch size are set for training. When testing, RMSE is used to measure accuracy.

[0081] After the new energy prediction deep learning model is trained, the power generation of new energy in the future can be predicted based on the trained new energy prediction deep learning model.

[0082] Since renewable energy generation is closely related to the local climate, and the climate is regular in a long-scale time series, we choose to use LSTM to build network predictions. At the same time, in order to ensure the accuracy and speed of the prediction, we use convolution plus bidirectional long short-term memory network to build the specific model. The convolution layer can better capture the sudden fluctuations and periodic peaks and troughs in the time series. In addition, since different renewable energy sources are affected differently by the climate, for example, photovoltaic power generation is mainly affected by sunlight and less by wind speed, while wind power generation is greatly affected by wind speed and has little to do with sunlight. Therefore, the renewable energy generators are grouped, and the corresponding neural network is trained separately for each group, which can further improve the accuracy of power generation predictions.

[0083] In some embodiments, the constraint conditions for constraining the first bidding coefficient of the new energy by the segmented bidding model of the new energy include:

[0084]

[0085] Among them, a i is the first quotation coefficient of the i-th quotation segment; P new,i is the actual power generation of renewable energy in the i-th quotation period; The predicted power generation of renewable energy in the i-th quotation period; the preset ratio is 0.9.

[0086] In some embodiments, the segmented quotation model for new energy sources further includes:

[0087] C new,i (P new,i )=a i P enw,i +b i (2)

[0088] y i P min,i ≤P new,i ≤y i P max,i (3)

[0089]

[0090] Among them, b i is the second quotation coefficient of the i-th quotation segment; C new,i (P new,i ) The price of new energy in the i-th quotation period; y i y is a 0-1 variable indicating whether the new energy output reported in the i-th quotation period is called, 1 means it is called, and 0 means it is not called; i+1 P is a 0-1 variable indicating whether the new energy in the i+1th quotation segment is called; min,i is the minimum power generation of renewable energy in the i-th quotation period; P max,i is the maximum power generation of renewable energy in the i-th quotation period; n is the number of quotation periods; P new is the total power generation of renewable energy.

[0091] The segmented bidding model of new energy in the embodiment of the present application introduces a series of 0-1 variables to couple the bidding segments, thereby converting it into a solvable mixed integer programming problem. The above equations (2)-(5) are respectively the bidding function of the i-th bidding segment, the output power limit of the i-th bidding segment, the output power continuous limit and the total output power limit.

[0092] After obtaining the predicted power generation of new energy in the current period in step 101, the first quotation coefficient a i There are additional constraints as shown in formula (1). Formula (1) makes the above-mentioned new energy segmented quotation model The lowest price at that time makes people more inclined to choose new energy when clearing out, while also ensuring a certain profit margin.

[0093] In order to increase the revenue of renewable energy and reduce the power abandonment rate, the embodiment of the present application introduces a segmented quotation model for renewable energy. This segmented quotation model quotes low prices in the low-power stage, slightly higher prices in the medium-power stage, and continues to increase prices in the high-power stage to prevent excessive electricity from being concentrated on renewable energy power generation and causing system instability.

[0094] In some embodiments, in step 103, based on the segmented quotation model of new energy, a SCUC clearing calculation model is determined, including:

[0095] Calculate the environmental costs of thermal power plants;

[0096] Based on the segmented bidding model of new energy and the environmental cost of thermal power units, the SCUC clearing calculation model is determined with the goal of minimizing cost.

[0097] The above formula for calculating the environmental cost of thermal power units is:

[0098]

[0099] Among them, C e,k is the environmental cost of the kth thermal power unit; is the fixed environmental cost of each startup of the kth thermal power unit; e,k P is the environmental cost generated by the kth thermal power unit for each unit of electricity produced, e,k is the actual output of the kth thermal power unit, and the actual output is the actual generated power.

[0100] After calculating the environmental cost of the thermal power unit, the embodiment of the present application combines the segmented quotation model of new energy, the environmental cost of the thermal power unit and the cost of the energy storage unit, etc., with the goal of minimizing cost, and can determine the SCUC clearing calculation model.

[0101] In some embodiments, the SCUC clearing calculation model includes a first objective function and a first constraint condition with the goal of minimizing cost; the first constraint condition includes a thermal power unit constraint, a line flow constraint, a node electricity price calculation constraint, and an energy storage device constraint.

[0102] In some embodiments, the first objective function includes:

[0103]

[0104] Among them, C e,k is the environmental cost of the kth thermal power unit; C k,t (P k,t ) is the operating cost of the kth thermal power unit in the tth period; is the startup cost of the kth thermal power unit in the tth period; is the discharge price of the es-th energy storage unit in the t-th period; is the charging price of the es-th energy storage unit in the t-th period; is the discharge power of the es-th energy storage unit in the t-th period; is the charging power of the es-th energy storage unit in the t-th period; C new,lt is the cost of the lth renewable energy unit in the tth period; β is the power abandonment penalty coefficient; is the power abandonment of the lth renewable energy unit in the tth period; min f is the first objective function.

[0105] C k,t (P k,t ) is calculated by referring to formula (8), The calculation formula of is shown in formula (9).

[0106]

[0107] in, is the electricity price reported by the kth thermal power unit in the tth time period and the ith quotation period, P k,t,i is the declared output of the kth thermal power unit in the tth time period and the ith quotation period, that is, the declared power generation power; μ k,t is a 0-1 variable indicating the start and stop of the kth thermal power unit in the tth period, is the startup cost of the kth thermal power unit; μ k,(t-1) It is a 0-1 variable indicating the start and stop of the k-th thermal power unit in the t-1th period.

[0108] In some embodiments, The calculation formula includes:

[0109]

[0110] in, is the predicted power generation of the lth new energy unit in the tth period, i.e. the predicted output; is the actual power generation of the lth new energy unit in the tth period.

[0111] For the power abandonment penalty coefficient β, the initial value is recommended to be set to 1.5 times the marginal cost of the new energy unit. Then, after each clearance, the value of β is dynamically adjusted according to the actual power abandonment situation of the new energy and the desired power abandonment rate. However, the value of β should not be too high. Too high a value may lead to problems such as high power generation pressure on the new energy units, reduced overall system stability and economic benefits.

[0112] The above-mentioned thermal power unit constraints may include equations (13)-(22).

[0113]

[0114] Among them, P k,t,i It can be the declared output of the kth thermal power unit in the ith quotation segment in the tth period, or it can be the winning bid output of the kth thermal power unit in the ith quotation segment in the tth period; is the maximum winning bid output of the kth thermal power unit in the ith bidding period in the tth period; P k,t is the total bid output of the kth thermal power unit in the tth period; μ k,t is a 0-1 variable indicating the start and stop of the k-th thermal power unit in the t-th period; is the minimum output of the kth thermal power unit in the tth period; is the maximum output of the kth thermal power unit in the tth period; is the discharge power of the es-th energy storage unit in the t-th period; is the charging power of the es-th energy storage unit in the t-th period; is the actual power generation of the lth renewable energy unit in the tth period; is the total system load in the tth period; r SR Plan the reserve ratio for the system; P k,t-1 is the total winning bid output of the kth thermal power unit in the t-1th period; is the climbing capacity of the kth thermal power unit in adjacent time periods; is the startup time of the kth thermal power unit in the tth period; μ k,t-1 is a 0-1 variable indicating the start and stop of the k-th thermal power unit in the t-1th period; is the minimum startup time of the kth thermal power unit; is the downtime of the kth thermal power unit in the tth period; is the minimum downtime of the kth thermal power unit; μ k,τ It is a 0-1 variable indicating the start and stop of the kth thermal power unit at time τ.

[0115] The above formulas (13)-(20) are respectively the bidding constraints of each bidding segment of the thermal power unit, the calculation of the total bidding output of the thermal power unit, the total bidding output constraint of the thermal power unit, the system load balance constraint, the system planned reserve constraint, the thermal power unit ramp constraint, the minimum start-up time constraint of the thermal power unit and the minimum shutdown time constraint of the thermal power unit.

[0116] The above-mentioned line power flow constraints may include equations (23)-(24).

[0117]

[0118] Among them, P bd,t is the line flow between line node b and line node d in the tth period; is the upper limit of the line power flow between line node b and line node d; x bd is the reactance between line node b and line node d; δ b,t is the phase angle of line node b in the tth period; δ d,t is the phase angle of line node d in the tth period.

[0119] The above node electricity price calculation constraint may include formula (25).

[0120]

[0121] When calculating the benefits of all parties in the market, it is necessary to calculate the node electricity price of the current line node. The node power balance constraint can be established or the power flow constraint can be established using the transfer distribution factor to perform the calculation. The formula is shown in formula (25). is the load of line node b in the tth period; b,t is the shadow price of the constraint, serving as the node electricity price of line node b in the tth period.

[0122] The energy storage device constraints may include equations (26)-(31), which are respectively energy storage unit discharge power constraints, energy storage unit charging power constraints, and SOC constraints for adjacent, initial, and final time periods.

[0123]

[0124]

[0125] in, is the maximum discharge power of the es-th energy storage unit in the t-th period; The maximum charging power of the es-th energy storage unit in the t-th period; E es,t is the SOC state of the es-th energy storage unit in the t-th period; E es,t-1 is the SOC state of the es-th energy storage unit in the t-1th period; is the charging efficiency of the esth energy storage unit; is the discharge efficiency of the es-th energy storage unit; is the minimum SOC state of the es-th energy storage unit; is the maximum SOC state of the es-th energy storage unit; E es,0 is the SOC state of the es-th energy storage unit in the 0th period; The initial SOC state of the esth energy storage unit; E es,T is the SOC state of the es-th energy storage unit in the T-th period; The final SOC state of the esth energy storage unit.

[0126] The charging and discharging state of energy storage can be determined by the relationship between the predicted output of new energy and the actual demand, which can be expressed as follows using the ternary operator:

[0127]

[0128] In formula (32), is the predicted power generation output of the new energy. Equation (32) shows that when the predicted power generation output is greater than the demand, the energy storage will be charged, and when the predicted power generation output is less than the demand, it will be discharged.

[0129] In some possible implementations, the first constraint condition may also include the above-mentioned equations (1)-(6) and (8)-(12).

[0130] In some embodiments, the SCED model includes a second objective function and a second constraint;

[0131] The second objective function includes:

[0132]

[0133] Among them, C k,t (P k,t ) is the operating cost of the kth thermal power unit in the tth period; is the discharge price of the es-th energy storage unit in the t-th period; is the charging price of the es-th energy storage unit in the t-th period; is the discharge power of the es-th energy storage unit in the t-th period; is the charging power of the es-th energy storage unit in the t-th period; C new,lt is the cost of the lth renewable energy unit in the tth period; β is the power abandonment penalty coefficient; is the power abandonment of the lth renewable energy unit in the tth period; min E is the second objective function.

[0134] The embodiment of the present application uses the SCUC algorithm to determine the winning power generation and energy storage units and uses SCED to determine the specific power generation output plan of the winning units, wherein a new energy power abandonment penalty mechanism is introduced, and a penalty term is added to the objective function to control the abandonment of new energy power; the embodiment of the present application can effectively improve the problem of wind and solar power abandonment of new energy, and at the same time has good solution efficiency.

[0135] The second constraint condition may include the above-mentioned equations (1)-(5), (8) and (10)-(32).

[0136] In view of the current problems of strong volatility of new energy power generation, difficult prediction of power generation, and insufficient support for new energy power generation by the power market clearing mechanism, the embodiment of the present application proposes a new energy prediction deep learning model combining convolutional neural network and bidirectional long short-term memory network. First, based on the close correlation between new energy power generation and climatic conditions, a convolution + bidirectional long short-term memory network model is constructed to improve the accuracy of power generation prediction and cope with power generation fluctuations. Secondly, a segmented quotation model for new energy is proposed. By introducing 0-1 variables to couple each quotation segment, the quotation problem is converted into mixed integer programming to optimize the market competitiveness of new energy power generation. Thirdly, the SCUC algorithm and the SCED algorithm are combined to determine the optimal combination and output plan of the generator set and the energy storage device, and a new energy abandonment penalty mechanism is introduced to reduce the abandonment rate of new energy. Through the peak-shifting and valley-filling function of energy storage, the embodiment of the present application improves the scheduling flexibility of the system, promotes the consumption of renewable energy, and reduces the abandonment of wind and light. In addition, the entire system comprehensively considers factors such as load balance, node electricity price, and energy storage charging and discharging constraints to ensure the economy and computational efficiency of market clearing.

[0137] In a specific application scenario, the above method may include:

[0138] Step 1: Build a new energy prediction deep learning model that combines convolutional neural networks and bidirectional long short-term memory networks. The model is trained for 100 rounds and 32 batches, and predictions are performed. Figure 2 Schematic diagram of the training of the deep learning model for new energy prediction and the prediction curve, where the horizontal axis is time (Datetime), the vertical axis is the power generation (Power), 21 is the actual curve (Actual), and 22 is the predicted curve (Predicted). Figure 3 This is a schematic diagram of using a trained new energy prediction deep learning model to predict the power generation of new energy in the next 24 hours.

[0139] Step 2: Read the relevant data of thermal power units, energy storage units, new energy units and system-related load data.

[0140] Step 3: Use a linear solver to define the first objective function of SCUC and related constraints, solve them, and store the results in a file, where the power abandonment penalty coefficient β is set to 1.5 times the marginal cost of renewable energy generation.

[0141] Step 4: Read the relevant result data of SCUC, use the linear solver to define the second objective function of SCED and related constraints, and solve the output plan.

[0142] Step 5: Calculate the benefits of each market party based on the results of SCED.

[0143] Based on the above steps, the simulation results are as follows Figure 4-Figure 9 shown. Figure 4 The load curve of the system in one day in the example is shown; Figure 5 The network parameters of the system in the example are shown; Figure 6 The figure shows the winning bids of thermal power units after the clearance. Figure 6 In the table, different Units represent different thermal power units; Figure 7 It shows the winning bids of energy storage units after the clearance. Figure 7 In the figure, different Es represent different energy storage units; Figure 8 It shows the winning bids of new energy units after the clearance. Figure 8 In the above, different Newgens represent different new energy units; Fig. 9 Shows the power abandonment rate of renewable energy.

[0144] According to the optimization results, in this example, in the SCUC stage, the new energy units won the bid in 96 time periods, and the thermal power units were mainly used as power supplements during peak power consumption. Negative winning power means that the energy storage units in this period purchased and stored electricity as buyers in the market, and released the electricity during peak power consumption. Therefore, the income of the storage period is also shown as negative in the income. Finally, in terms of the abandonment of new energy, the abandonment rate is high when the power consumption is low, but the abandonment rate will drop overall during peak periods. The overall abandonment rate is about 8%, which can well achieve the goal of limited consumption of new energy.

[0145] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0146] The following is an embodiment of the device of the present invention. For details not described in detail therein, reference may be made to the corresponding method embodiment described above.

[0147] Fig.10 The following is a schematic diagram showing the structure of a smart grid dispatching device with independent energy storage and new energy provided by an embodiment of the present invention. For ease of explanation, only the parts related to the embodiment of the present invention are shown, which are described in detail as follows:

[0148] like Fig.10 As shown, the smart grid dispatching device 100 with independent energy storage and new energy participation includes: a prediction module 101, a new energy quotation model determination module 102, a first solution module 103, a second solution module 104 and a dispatching module 105.

[0149] The prediction module 101 is used to predict the power generation of the new energy and obtain the predicted power generation of the new energy;

[0150] The new energy quotation model determination module 102 is used to determine the segmented quotation model of the new energy based on the predicted power generation of the new energy; the segmented quotation model of the new energy constrains the first quotation coefficient of the new energy so that when the ratio of the actual power generation of the new energy to the corresponding predicted power generation is a preset ratio, the quotation of the new energy is the lowest;

[0151] A first solution module 103 is used to determine a SCUC clearing calculation model based on a segmented quotation model of new energy, and solve the SCUC clearing calculation model to obtain a combination of power generation and energy storage units;

[0152] The second solution module 104 is used to solve the power generation plan and node electricity price based on the combination of power generation and energy storage units and a predetermined SCED model;

[0153] The scheduling module 105 is used to perform grid scheduling based on the power generation plan and node electricity prices.

[0154] In a possible implementation, the constraint conditions for constraining the first bidding coefficient of the new energy by the segmented bidding model of the new energy include:

[0155]

[0156] Among them, a i is the first quotation coefficient of the i-th quotation segment; P new,i is the actual power generation of renewable energy in the i-th quotation period; The predicted power generation of renewable energy in the i-th quotation period; the preset ratio is 0.9.

[0157] In a possible implementation, the segmented quotation model for new energy also includes:

[0158] C new,i (P new,i )=a i P new,i +b i

[0159] y i P min,i ≤P new,i ≤y i P max,i

[0160]

[0161] Among them, b i is the second quotation coefficient of the i-th quotation segment; C new,i (P new,i ) The price of new energy in the i-th quotation period; y iP is a 0-1 variable indicating whether the new energy output reported in the i-th quotation period is called; min,i is the minimum power generation of renewable energy in the i-th quotation period; P max,i is the maximum power generation of renewable energy in the i-th quotation period; n is the number of quotation periods; P new is the total power generation of renewable energy.

[0162] In a possible implementation, in the first solution module 113, based on the segmented quotation model of new energy, the SCUC clearing calculation model is determined, including:

[0163] Calculate the environmental costs of thermal power plants;

[0164] Based on the segmented bidding model of new energy and the environmental cost of thermal power units, the SCUC clearing calculation model is determined with the goal of minimizing cost.

[0165] In one possible implementation, the SCUC clearing calculation model includes a first objective function with the lowest cost as the goal and a first constraint condition; the first constraint condition includes a thermal power unit constraint, a line flow constraint, a node electricity price calculation constraint, and an energy storage device constraint.

[0166] In a possible implementation, the first objective function includes:

[0167]

[0168] Among them, C e,k is the environmental cost of the kth thermal power unit; C k,t (P k,t ) is the operating cost of the kth thermal power unit in the tth period; is the electricity price reported by the kth thermal power unit in the ith quotation period of the tth period, P k,t,i The declared output of the kth thermal power unit in the ith quotation period during the tth period; is the startup cost of the kth thermal power unit in the tth period; μ k,t is a 0-1 variable indicating the start and stop of the kth thermal power unit in the tth period, is the startup cost of the kth thermal power unit; is the discharge price of the es-th energy storage unit in the t-th period; is the charging price of the es-th energy storage unit in the t-th period; is the discharge power of the es-th energy storage unit in the t-th period; is the charging power of the es-th energy storage unit in the t-th period; C new,lt is the cost of the lth renewable energy unit in the tth period; β is the power abandonment penalty coefficient; is the power abandonment of the lth renewable energy unit in the tth period; min f is the first objective function.

[0169] In one possible implementation,

[0170] in, is the predicted power generation of the lth new energy unit in the tth period; is the actual power generation of the lth new energy unit in the tth period.

[0171] In a possible implementation, the SCED model includes a second objective function and a second constraint;

[0172] The second objective function includes:

[0173]

[0174] Among them, C k,t (P k,t ) is the operating cost of the kth thermal power unit in the tth period; is the discharge price of the es-th energy storage unit in the t-th period; is the charging price of the es-th energy storage unit in the t-th period; is the discharge power of the es-th energy storage unit in the t-th period; is the charging power of the es-th energy storage unit in the t-th period; C new,lt is the cost of the lth renewable energy unit in the tth period; β is the power abandonment penalty coefficient; is the power abandonment of the lth renewable energy unit in the tth period; min E is the second objective function.

[0175] In a possible implementation, the prediction module 111 is specifically configured to:

[0176] Based on the pre-trained new energy prediction deep learning model, the power generation of new energy is predicted to obtain the predicted power generation of new energy; the new energy prediction deep learning model includes convolution layer, pooling layer, bidirectional long short-term memory layer and fully connected layer.

[0177] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0178] Those of ordinary skill in the art will appreciate that the templates, units, and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0179] If the module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, it can implement the steps of the above-mentioned smart grid dispatching method embodiments involving each independent energy storage and new energy. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory, random access memory, electrical carrier signal, telecommunication signal and software distribution medium.

[0180] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A smart grid dispatching method involving independent energy storage and new energy, characterized in that: include: Predict the power generation of new energy to obtain the predicted power generation of new energy; Based on the predicted power generation of the new energy, a segmented quotation model of the new energy is determined; the segmented quotation model of the new energy constrains the first quotation coefficient of the new energy, so that when the ratio of the actual power generation of the new energy to the corresponding predicted power generation is a preset ratio, the quotation of the new energy is the lowest; Based on the segmented bidding model of the new energy, a SCUC clearing calculation model is determined, and the SCUC clearing calculation model is solved to obtain a combination of power generation and energy storage units; Based on the combination of the power generation and energy storage units and a predetermined SCED model, a power generation plan and a node electricity price are obtained; The power grid is dispatched based on the power generation plan and the node electricity price.

2. The method for dispatching a smart grid with independent energy storage and new energy according to claim 1, characterized in that: The constraint conditions for constraining the first quotation coefficient of the new energy by the segmented quotation model of the new energy include: Among them, a i is the first quotation coefficient of the i-th quotation segment; new,i is the actual power generation of renewable energy in the i-th quotation period; is the predicted power generation of the renewable energy in the i-th quotation period; the preset ratio is 0.

9.

3. The smart grid dispatching method involving independent energy storage and new energy according to claim 2 is characterized in that: The segmented quotation model of new energy also includes: C new,i (P new,i )=aiP new,i +b i y i P min,i ≤P new,i ≤y i P max,i Among them, b i is the second quotation coefficient of the i-th quotation segment; C new,i (P new,i ) The price of new energy in the i-th quotation period; y i P is a 0-1 variable indicating whether the new energy output reported in the i-th quotation period is called; min,i is the minimum power generation of renewable energy in the i-th quotation period; P max,i is the maximum power generation of renewable energy in the i-th quotation period; n is the number of quotation periods; P new is the total power generation of renewable energy.

4. The method for dispatching a smart grid with independent energy storage and new energy according to claim 1, characterized in that: The SCUC clearing calculation model is determined based on the segmented quotation model of the new energy, including: Calculate the environmental costs of thermal power plants; Based on the segmented quotation model of the new energy and the environmental cost of the thermal power unit, the SCUC clearing calculation model is determined with the lowest cost as the goal.

5. The method for dispatching a smart grid with independent energy storage and new energy according to claim 1, characterized in that: The SCUC clearing calculation model includes a first objective function with the lowest cost as the goal and a first constraint condition; the first constraint condition includes a thermal power unit constraint, a line flow constraint, a node electricity price calculation constraint and an energy storage device constraint.

6. The method for dispatching a smart grid with independent energy storage and new energy as claimed in claim 5, characterized in that: The first objective function includes: Among them, C e,k is the environmental cost of the kth thermal power unit; C k,t (P k,t ) is the operating cost of the kth thermal power unit in the tth period; is the electricity price reported by the kth thermal power unit in the tth time period and the ith quotation period, P k,t,i The declared output of the kth thermal power unit in the ith quotation period during the tth period; is the startup cost of the kth thermal power unit in the tth period; μ k,t is a 0-1 variable indicating the start and stop of the kth thermal power unit in the tth period, is the startup cost of the kth thermal power unit; is the discharge price of the es-th energy storage unit in the t-th period; is the charging price of the es-th energy storage unit in the t-th period; is the discharge power of the es-th energy storage unit in the t-th period; is the charging power of the es-th energy storage unit in the t-th period; C new,lt is the cost of the lth renewable energy unit in the tth period; β is the power abandonment penalty coefficient; is the power abandonment of the lth new energy unit in the tth time period; min f is the first objective function.

7. The method for dispatching a smart grid with independent energy storage and new energy as claimed in claim 6, characterized in that: in, is the predicted power generation of the lth new energy unit in the tth period; is the actual power generation of the lth new energy unit in the tth period.

8. The method for dispatching a smart grid with independent energy storage and new energy as claimed in any one of claims 1 to 7, characterized in that: The SCED model includes a second objective function and a second constraint condition; The second objective function includes: Among them, C k,t (P k,t ) is the operating cost of the kth thermal power unit in the tth period; is the discharge price of the es-th energy storage unit in the t-th period; is the charging price of the es-th energy storage unit in the t-th period; is the discharge power of the es-th energy storage unit in the t-th period; is the charging power of the es-th energy storage unit in the t-th period; C new,lt is the cost of the lth renewable energy unit in the tth period; β is the power abandonment penalty coefficient; is the abandoned power of the lth new energy unit in the tth time period; min E is the second objective function.

9. The method for dispatching a smart grid with independent energy storage and new energy as claimed in any one of claims 1 to 7, characterized in that: The step of predicting the power generation of the new energy to obtain the predicted power generation of the new energy includes: Based on a pre-trained new energy prediction deep learning model, the power generation power of the new energy is predicted to obtain the predicted power generation power of the new energy; the new energy prediction deep learning model includes a convolutional layer, a pooling layer, a bidirectional long short-term memory layer and a fully connected layer.

10. A smart grid dispatching device with independent energy storage and new energy participation, characterized in that: include: A prediction module is used to predict the power generation of new energy and obtain the predicted power generation of new energy; A new energy quotation model determination module is used to determine a segmented quotation model of new energy based on the predicted power generation of the new energy; the segmented quotation model of new energy constrains the first quotation coefficient of the new energy so that when the ratio of the actual power generation of the new energy to the corresponding predicted power generation is a preset ratio, the quotation of the new energy is the lowest; A first solution module is used to determine a SCUC clearing calculation model based on the segmented quotation model of the new energy, and solve the SCUC clearing calculation model to obtain a combination of power generation and energy storage units; A second solution module is used to solve the power generation plan and node electricity price based on the combination of the power generation and energy storage units and a predetermined SCED model; The scheduling module is used to perform power grid scheduling based on the power generation plan and node electricity prices.

Images