A scheduling method, device, equipment, medium and product for grid power equipment

Abstract

The present invention discloses a scheduling method, device, equipment, medium and product for grid power equipment. The method includes: in response to a scheduling request for grid power equipment, clustering the historical prediction error data of the distribution network based on a preset density clustering algorithm to construct a prediction error scenario set according to the clustering result; determining the constraint conditions of the two-layer scheduling model according to the prediction error scenario set to construct the two-layer scheduling model; and solving the two-layer scheduling model based on a preset solver to obtain the target value of the grid power equipment for scheduling the grid power equipment. The technical solution of the present invention can comprehensively analyze from both the distribution network side and the industrial park microgrid side, establish an accurate scheduling model, determine the accurate scheduling value, realize the optimal scheduling of the grid power equipment, and thus achieve the effect of energy conservation and carbon reduction on the basis of taking into account the economic operation and safety of enterprises in the distribution network and the industrial park microgrid.

Description

Technical Field

[0001] The present invention relates to the field of distribution networks, and in particular, to a scheduling method, device, equipment, medium and product for grid power equipment. Background Art

[0002] With the development of the low-carbon transformation of energy under the dual-carbon goal, it is imperative to connect renewable energy to the distribution network on a large scale. In order to adapt to the uncertainty of renewable energy generation, it is necessary to explore from multiple perspectives of source, load and storage, so as to establish a relevant distribution network optimal scheduling model and flexibly adjust resources. At the same time, with the development of related theories such as carbon emission flow, the carbon emissions of industrial park enterprises should be accounted for on the enterprise side. Therefore, it is also necessary to establish an optimal scheduling model for industrial park microgrids considering carbon emission costs and participation in demand response accordingly.

[0003] Therefore, how to conduct a comprehensive analysis from both the distribution network side and the industrial park microgrid side, establish an accurate scheduling model, determine the accurate scheduling value, and realize the optimal scheduling of grid power equipment is an urgent problem to be solved at present. Summary of the Invention

[0004] The present invention provides a scheduling method, device, equipment, medium and product for grid power equipment, so as to conduct a comprehensive analysis from both the distribution network side and the industrial park microgrid side, establish an accurate scheduling model, determine the accurate scheduling value, and realize the optimal scheduling of grid power equipment.

[0005] According to one aspect of the present invention, a scheduling method for grid power equipment is provided, including:

[0006] In response to a scheduling request for grid power equipment, based on a preset density clustering algorithm, cluster the historical prediction error data of the distribution network to construct a prediction error scenario set according to the clustering result;

[0007] According to the prediction error scenario set, determine the constraint conditions of the two-layer scheduling model to construct the two-layer scheduling model; wherein, the two-layer scheduling model includes an upper-layer model and a lower-layer model, the upper-layer model is a distribution network side model, and the lower-layer model is an industrial park microgrid side model;

[0008] Based on a preset solver, solve the two-layer scheduling model to obtain the target value of the grid power equipment, so as to schedule the grid power equipment.

[0009] According to another aspect of the present invention, a scheduling device for grid power equipment is provided, including:

[0010] A clustering module, configured to, in response to a scheduling request for grid power equipment, based on a preset density clustering algorithm, cluster the historical prediction error data of the distribution network to construct a prediction error scenario set according to the clustering result;

[0011] A building block for determining the constraints of a two - layer scheduling model according to a set of prediction error scenarios to construct a two - layer scheduling model; wherein the two - layer scheduling model includes an upper - layer model and a lower - layer model, the upper - layer model is a distribution network - side model, and the lower - layer model is an industrial park micro - grid - side model;

[0012] A scheduling module for solving the two - layer scheduling model based on a preset solver to obtain the target values of grid power equipment for scheduling the grid power equipment.

[0013] According to another aspect of the present invention, there is provided an electronic device, the electronic device includes:

[0014] At least one processor; and

[0015] A memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the scheduling method of the grid power equipment according to any embodiment of the present invention.

[0017] According to another aspect of the present invention, there is provided a computer - readable storage medium storing computer instructions for causing a processor to implement the scheduling method of the grid power equipment according to any embodiment of the present invention when executed.

[0018] According to another aspect of the present invention, there is also provided a computer program product including a computer program, and the computer program implements the scheduling method of the grid power equipment according to any embodiment of the present invention when executed by a processor.

[0019] The technical solution of the embodiment of the present invention, in response to a scheduling request for grid power equipment, clusters the historical prediction error data of the distribution network based on a preset density clustering algorithm to construct a set of prediction error scenarios according to the clustering results; determines the constraints of the two - layer scheduling model according to the set of prediction error scenarios to construct the two - layer scheduling model; and solves the two - layer scheduling model based on a preset solver to obtain the target values of grid power equipment for scheduling the grid power equipment. By comprehensively analyzing from both the distribution network side and the industrial park micro - grid side, an accurate scheduling model can be established, accurate scheduling values can be determined, and the optimal scheduling of grid power equipment can be realized, so as to achieve the effect of energy conservation and carbon reduction on the basis of taking into account the economic efficiency and safety of the operation of enterprises in the distribution network and the industrial park micro - grid.

[0020] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for use in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0022] Figure 1 is a flowchart of a scheduling method for a power grid power device provided in Embodiment 1 of the present invention;

[0023] Figure 2 is a structural diagram of a scheduling device for a power grid power device provided in Embodiment 2 of the present invention;

[0024] Figure 3 is a schematic structural diagram of an electronic device provided in Embodiment 3 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0025] In order to enable those skilled in the art to better understand the solutions of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0026] It should be noted that the terms "first", "second", "target", "candidate", "alternative", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those clearly listed steps or units, but may include other steps or units that are not clearly listed or are inherent to these processes, methods, products, or devices. The acquisition, storage, use, processing, etc. of data in the technical solutions of this application all comply with the relevant regulations of national laws and regulations.

[0027] Embodiment 1

[0028] Figure 1 It is a flowchart of a scheduling method for grid power equipment provided in the first embodiment of the present invention; this embodiment is applicable to the situation of predicting the operating conditions of a distribution network and an industrial park microgrid based on a two-layer scheduling model to effectively schedule grid power equipment. This method can be executed by a scheduling device for grid power equipment, and the scheduling device for grid power equipment can be implemented in the form of hardware and / or software. The scheduling device for grid power equipment can be configured in an electronic device, such as the total control device of a distribution network and a workbench park microgrid, such as Figure 1 As shown, the scheduling method for grid power equipment includes:

[0029] S101. In response to a scheduling request for grid power equipment, based on a preset density clustering algorithm, cluster the historical prediction error data of the distribution network to construct a prediction error scenario set according to the clustering result.

[0030] Among them, the scheduling request refers to a request for adjusting the start-stop state and power size of grid power equipment. The preset density clustering algorithm can be, for example, the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) method. The historical prediction error data refers to the historical prediction error data stored in the distribution network, specifically, the prediction error data of new energy power generation output and load power pushed back at least 1 year from the scheduling date. The historical prediction error data can include the prediction error data generated by predicting each grid power equipment during the operation of the distribution network, and can also include the prediction error data obtained by comparing the load curve sent by the industrial park microgrid with the expected value by the distribution network. The prediction error scenario set refers to a data set that can characterize the distribution of prediction errors. The prediction error scenario set can specifically include the clustering centers corresponding to various historical prediction error data.

[0031] Optionally, based on a preset density clustering algorithm, cluster the historical prediction error data of the distribution network to construct a prediction error scenario set, including: based on a preset density clustering algorithm, cluster the historical prediction error data of the distribution network to divide the historical prediction error data into at least two categories; for each divided category, based on the center clustering algorithm, determine the target clustering center from the target data points included in each category, and generate a prediction error scenario set according to the target clustering centers corresponding to each divided category. Among them, the number of target clustering centers can be at least two.

[0032] Optionally, the prediction error historical data can be preprocessed based on the DBSCAN density clustering method to determine the number of clusters in the data. Further, for each class, based on the center clustering algorithm, the sum of the squares of the distances between any point in each class and all other points in the class is calculated, and the data point with the smallest sum of squares is selected as the clustering center of the corresponding cluster to obtain the clustering centers corresponding to each class; finally, the set of clustering centers corresponding to each class is determined as the prediction error scenario set.

[0033] For example, the historical data of the prediction error can be recorded as a data set containing n data points, that is, X = {x 1 ,…,x i ,…,x n}, then, the result of clustering X using DBSCAN density clustering algorithm is as follows:

[0034] X={C 1 , …, C s ,…,Cn clu} (1)

[0035] Among them, n clu C is the total number of clusters obtained by clustering. Specifically, the number of clusters can be directly determined as the number of cluster targets contained in the historical data of the prediction error, that is, the total number of classes. 1 ,C s ,Cn clu The first, s and n clu The set of data points in a class.

[0036] For example, since the DBSCAN algorithm can only obtain classification results and cannot find the cluster centers of each class, the central clustering algorithm can be used to determine the cluster center of each class. Specifically, the Euclidean distance can be selected as the standard for evaluating the similarity and distance between a data point and its class, and the sum of the squares of the distances between any point in each class and other data points in the class is calculated, that is,

[0037]

[0038] Among them, x i and x j Both C s The data points are C s The i-th and j-th data points in J(x i ) is x i With C s The sum of the squares of the distances between other data points in the .

[0039] For example, for each class, the data point corresponding to the minimum sum of squares can be selected as the cluster center of the class, and so on. cluFor each class, perform the above calculations of the sum of squares and the selection of the clustering center, and the target clustering center corresponding to each class can be obtained. Finally, the set composed of the target clustering centers corresponding to each class can be determined as the prediction error scenario set.

[0040] Exemplarily, the prediction error scenario set can be expressed as:

[0041]

[0042] Among them, X typ is the typical scenario set of the source load day-ahead prediction error. are the clustering centers of C 1 , C s and Cn clu respectively. Taking as an example, the elements that each clustering center can contain are as follows:

[0043]

[0044] Among them, ΔW 1,1,s represents the prediction error of the output power of the first new energy power generation station in the first time period in the s-th typical scenario; ΔW j,t,s represents the prediction error of the output power of the j-th new energy power generation station in the t-th time period in the s-th typical scenario; represents the prediction error of the n w -th new energy power generation station in the T N -th time period in the s-th typical scenario; T N is the total number of time periods in a day; n w is the total number of new energy power generation stations in the distribution network to be optimized; ΔL 1,1,s represents the prediction error of the load power of the first load node in the first time period in the s-th typical scenario; ΔL d,t,s represents the prediction error of the output power of the d-th load node in the t-th time period in the s-th typical scenario; represents the load power prediction error of the n D -th load node in the T N -th time period in the s-th typical scenario; n D is the total number of load nodes in the distribution network to be optimized.

[0045] S102. Determine the constraint conditions of the two-layer scheduling model according to the prediction error scenario set to construct the two-layer scheduling model.

[0046] Among them, the two-layer scheduling model includes an upper-layer model and a lower-layer model. The upper-layer model is the distribution network side model, and the lower-layer model is the industrial park microgrid side model. The upper-layer model is essentially a power supply side model, and the lower-layer model is essentially a user side model. By optimizing and solving the two-layer scheduling model, the target values of the grid power equipment can be obtained, and thus the scheduling adjustment can be carried out.

[0047] Optionally, according to the constraint rules in the baseline scenario, the setting values of the constant parameters in the constraint conditions can be determined, and the constraint conditions in the baseline scenario can be constructed based on the setting values. The constant parameters can be, for example, the product output that must be completed every day in the output constraint, etc. And according to the data information in the prediction error scenario set, the constraint conditions in the prediction error scenario can be constructed to determine the constraint conditions of the two-layer scheduling model.

[0048] Optionally, the constraint conditions of the two-layer scheduling model include the constraint conditions in the baseline scenario and the constraint conditions in the prediction error scenario; the constraint conditions in the baseline scenario include at least one of the following: power balance constraint, gas turbine output constraint, gas turbine ramping constraint, gas turbine minimum start-stop time constraint, gas turbine reserve capacity constraint, and branch power flow constraint; the constraint conditions in the prediction error scenario include at least one of the following: scenario energy balance constraint, scenario unit ramping constraint, scenario branch power flow constraint, unit up and down reserve capacity call constraint, new energy curtailment power constraint, load shedding power constraint, and enterprise constraints of electrolytic and heat storage enterprises in the industrial park microgrid. For example, the constraint conditions in the baseline scenario can be specifically referred to in formulas (32)-(37), and the constraint conditions in the prediction error scenario can be specifically referred to in formulas (38)-(43).

[0049] Optionally, the enterprise constraints of electrolytic and heat storage enterprises in the industrial park microgrid include at least one of the following: electrolytic equipment power constraint, electrolytic equipment reserve capacity constraint, electrolytic equipment temperature constraint, electrolytic enterprise daily output constraint, electrolytic equipment reserve capacity actual call constraint, power energy constraint corresponding to electrolytic equipment, heat storage load daily regulation capacity constraint, heat storage industrial load daily regulation times constraint, daily output constraint of heat storage industrial load, and power energy constraint corresponding to heat storage equipment. Exemplarily, the enterprise constraints can be specifically referred to in formulas (5)-(19).

[0050] Exemplarily, the electrolytic equipment power constraint can be expressed by the following formula:

[0051]

[0052] where the subscripts l and t are the numbers of the electrolytic equipment and the time period respectively; P l,t is the power of the electrolytic equipment l at time period t; P l max and P l min are the maximum and minimum values of the power of the electrolytic equipment l respectively; U l is the voltage of the electrolytic equipment l; I l,t is the current intensity passed through the electrolytic equipment l at time period t.

[0053] Exemplarily, the spare capacity constraint of the electrolytic device can be expressed by the following formula:

[0054]

[0055] where V l,t is the spare capacity reserved for electrolytic device l at time period t; P l,t is the power of electrolytic device l at time period t; represents the current that needs to be reduced for electrolytic device l to reserve spare capacity at time period t. P l min is the minimum power of electrolytic device l; U l is the voltage of electrolytic device l.

[0056] Exemplarily, the temperature constraint of the electrolytic device can be expressed by the following formula:

[0057] T e,l,1,s = T e max (7)

[0058]

[0059]

[0060] where T e,l,1,s 、T e,l,2,s 、T e,l,t-1,s 、T e,l,t-2,s and T e,l,t,s respectively represent the electrolyte temperatures of electrolytic device l at time periods 1, 2, t - 1, t - 2, and t under scenario s; I l,t-1 and I l,t-2 are the current intensities passed through electrolytic device l at time periods t - 1 and t - 2; and represent the currents actually reduced due to the reserved spare capacity of electrolytic device l at time periods 1, t - 1, and t - 2 under scenario s; a and b are the inertia coefficients of the current temperature with respect to the temperatures of the previous two time periods; c and d are the inertia coefficients of the current temperature with respect to the currents of the previous two time periods; K is the temperature constant related to the internal structure of the electrolytic device; I max is the maximum current allowed to pass through the electrolytic device; T e min and T e max respectively represent the minimum and maximum allowable values of the electrolyte temperature.

[0061] Exemplarily, the daily output constraint of the electrolytic enterprise can be expressed by the following formula:

[0062]

[0063] Among them, and represent the minimum and maximum daily energy consumption of the electrolytic equipment l, which are directly proportional to the minimum and maximum product outputs respectively. I l,t is the current intensity passed through the electrolytic equipment l during the time period t. represents the current actually reduced when the electrolytic equipment l is called for standby capacity during the time period t under the scenario s; T N is the total number of time periods in a day; U l is the voltage of the electrolytic equipment l.

[0064] Exemplarily, the actual call constraint for the standby capacity of the electrolytic equipment can be expressed as:

[0065] 0 ≤ V l,t,s ≤ V l,t , (12)

[0066] Among them, V l,t,s is the size of the actually called standby capacity of the electrolytic equipment l during the time period t under the scenario s. V l,t is the reserved standby capacity of the electrolytic equipment l during the time period t.

[0067] Exemplarily, the power and energy constraints of the energy storage device installed at the electrolytic equipment, that is, the power and energy constraints corresponding to the electrolytic equipment can be expressed as:

[0068]

[0069] Among them, are the charge and discharge status indicator variables of the energy storage device installed at the l-th electrolytic equipment during the time period t under the scenario s, both of which are 0-1 variables; and are the charge and discharge powers of the energy storage device installed at the electrolytic equipment l during the time period t; η l are the maximum charge and discharge power, the maximum energy storage capacity and the charge and discharge efficiency of the energy storage device installed at the l-th electrolytic equipment; is the real-time stored energy of the energy storage device installed at the l-th electrolytic equipment during the time period t under the scenario s; S lmin 、S lmax are the minimum and maximum state of charge coefficients of the energy storage device installed at the l-th electrolytic equipment respectively.

[0070] Exemplarily, the daily regulation capacity constraint of the heat storage type load can be expressed as:

[0071]

[0072] Among them, P m,t,sis the operating power of the m-th heat storage type load after regulation in period t under scenario s. is the reference power of the m-th heat storage type industrial load operation. are the optimization variables corresponding to the up-regulation and down-regulation powers of the m-th heat storage type industrial load in period t under scenario s, su m,t,s and sd m,t,s are 0-1 variables reflecting the states of power up-regulation and down-regulation of the m-th heat storage type industrial load equipment in period t under scenario s. are the maximum up-regulation and down-regulation powers of the heat storage type industrial load on the premise of ensuring safe operation.

[0073] Exemplarily, the daily regulation times constraint of the heat storage type industrial load can be expressed as:

[0074]

[0075] where y m,t,s and z m,t,s are respectively the state conversion variables of power up-regulation and down-regulation of the m-th heat storage type industrial load equipment in period t under scenario s; that is, when y m,t,s and z m,t,s take 1, it represents that the m-th heat storage type industrial load equipment changes from other states to the power up-(down)-regulation state under scenario s; when taking 0, it represents that such a state change does not occur; M is the maximum set regulation times of the heat storage type industrial load; su m,t-1,s is a 0-1 variable representing that the m-th heat storage type industrial load equipment is in the power up-regulation state in period t-1 under scenario s.

[0076] Exemplarily, the daily output constraint of the heat storage type industrial load can be expressed as:

[0077]

[0078] where O m,t,s is the output of the m-th heat storage type industrial equipment in period t under scenario s; O m is the target output value of the heat storage type industrial load; λ 1 and λ 2 and λ 3 are respectively the product yields of the heat storage type industrial load in the power down-regulation state, rated power state, and up-regulation state. is the reference power of the m-th heat storage type industrial load operation, sd m,t,s is a 0-1 variable reflecting the states of power up-regulation and down-regulation of the m-th heat storage type industrial load equipment in period t under scenario s; su m,t-1,s is a 0-1 variable representing that the m-th heat storage type industrial load equipment is in the power up-regulation state in period t-1 under scenario s. The maximum upward and downward regulation powers of the heat storage type industrial load under the premise of ensuring safe operation.

[0079] Exemplarily, the power and energy constraints of the energy storage device installed at the heat storage type device, that is, the power and energy constraints corresponding to the heat storage type device can be expressed as:

[0080]

[0081] Among them, are the charge and discharge state indication variables of the energy storage device installed at the m-th heat storage type industrial load device in period t under scenario s, both being 0-1 variables; and are the charge and discharge powers of the energy storage device installed at the heat storage type device m in period t; η m are the maximum charge and discharge power, the maximum energy storage capacity, and the charge and discharge efficiency of the energy storage device installed at the m-th heat storage type device; is the real-time stored energy of the energy storage device installed at the m-th heat storage type industrial load device in period t under scenario s; S mmin 、S mmax are the minimum and maximum state of charge coefficients of the energy storage device installed at the m-th heat storage type industrial load device respectively.

[0082] Exemplarily, the lower layer model in the two-layer scheduling model, that is, the industrial park microgrid side model, can be specifically expressed as a low-carbon optimal scheduling model for the coordinated operation of the source, load, and storage in the industrial park with multiple types of enterprises. Its constraint conditions are formulas (5)-(19), and the objective function is to maximize the total net income of all electrolytic and heat storage type industrial enterprises in the industrial park.

[0083] Optionally, the objective function corresponding to the industrial park microgrid side model can be constructed through the net income objective function obtained by electrolytic enterprises and the net income objective function obtained by heat storage type enterprises. Among them, the net income objective function obtained by electrolytic enterprises includes multiple consideration indicators such as product income, income from participating in demand response by enterprises, operation and maintenance costs, and carbon emission costs; the net income objective function obtained by heat storage type enterprises includes multiple consideration indicators such as product production income, income from participating in demand response, and carbon emission costs.

[0084] Exemplarily, the product income item in the net income objective function obtained by electrolytic enterprises can be expressed as:

[0085] A l,t =α l U l I l,t

[0086]

[0087] Among them, A l,t is the profit obtained from product production by the electrolytic equipment l during the time period t; A l,t,s is the product profit loss caused by the actually called standby power of the electrolytic equipment l during the time period t; α l is the product production profit price coefficient of the electrolytic equipment l, in which factors such as raw material cost and electricity cost have been considered, and this price coefficient can be obtained by converting the profit per ton of product. U l is the voltage of the electrolytic equipment l. I l,t is the current intensity passed through the electrolytic equipment l during the time period t. represents the current actually reduced when the electrolytic equipment l is called for standby capacity during the time period t under the scenario s.

[0088] Exemplarily, the revenue of the electrolytic enterprise participating in demand response in the net revenue objective function can be expressed as:

[0089]

[0090] Among them, is the energy compensation provided by the power system operator to the electrolytic equipment l according to the response of the electrolytic enterprise load; is the energy compensation price; is the standby capacity compensation provided to the electrolytic equipment l; is the standby capacity compensation price; is the standby deployment compensation provided to the electrolytic equipment l during the time period t under the scenario s; is the standby deployment compensation price. U l is the voltage of the electrolytic equipment l. I l,t is the current intensity passed through the electrolytic equipment l during the time period t. represents the current actually reduced when the electrolytic equipment l is called for standby capacity during the time period t under the scenario s. represents the current that needs to be reduced for reserve standby of the electrolytic equipment l during the time period t.

[0091] Exemplarily, the operation and maintenance cost in the net revenue objective function obtained by the electrolytic enterprise can be expressed by the following formula:

[0092]

[0093] Among them, is the cost required for equipment maintenance when the electrolytic equipment l provides demand response during the time period t; is the additional equipment maintenance cost when the electrolytic equipment l is actually called for standby capacity during the time period t under the scenario s; c o&m is the cost coefficient of the electrolytic equipment load regulation.

[0094] Exemplarily, the carbon emission cost of the electrolytic enterprise in the net income objective function obtained by the electrolytic enterprise can be expressed by the following formula:

[0095]

[0096] Wherein, is the carbon emission cost of the electrolytic equipment l at time t; c carbon is the unit carbon emission cost coefficient; η AL is the unit power carbon emission factor of the internal process of the electrolytic equipment; is the dynamic unit power carbon emission factor of power consumption, which is related to the proportion of fossil fuel power generation in the power grid where the electrolytic enterprise is located; η AL,p is the unit power carbon emission quota of the electrolytic equipment.

[0097] In summary, the net income objective function obtained by the electrolytic enterprise can be expressed as:

[0098]

[0099] Wherein, B AL is the net income obtained by the electrolytic enterprise participating in demand response; N AL is the total number of electrolytic equipment in the industrial park; p s is the probability of the occurrence of scenario s. is the energy compensation provided by the power system operator to the electrolytic equipment l according to the response of the electrolytic enterprise load; is the reserve capacity compensation provided to the electrolytic equipment l; is the cost of equipment maintenance required for the electrolytic equipment l to provide demand response at time t; is the carbon emission cost of the electrolytic equipment l at time t; A l,t,s is the product profit loss caused by the actual call of the standby power of the electrolytic equipment l at time t; is the standby deployment compensation provided to the electrolytic equipment l at time t under scenario s; is the additional equipment maintenance cost when the electrolytic equipment l is actually called for standby capacity at time t under scenario s; T N is the total number of time periods in a day; N s is the number of scenarios.

[0100] Exemplarily, the product production income item in the net income objective function obtained by the heat storage enterprise can be expressed as:

[0101] A m,t,s = α m O m,t,s (27)

[0102] Among them, A m,t,s is the product profit loss caused by the actually called standby power of the heat storage type industrial equipment m during the time period t; α m is the product production profit price coefficient of the heat storage type industrial equipment m, O m,t,s is the output of the m-th heat storage type industrial equipment during the time period t under the scenario s.

[0103] Exemplarily, the revenue from participating in demand response in the net revenue objective function obtained by the heat storage type enterprise can be expressed as:

[0104]

[0105] Among them, is the revenue from the m-th heat storage type equipment participating in demand response during the time period t under the scenario s; c m1 , c m2 are respectively the unit response power subsidies for a single upward and downward adjustment of the heat storage type equipment; are the unit upward and downward adjustment powers of the m-th heat storage type equipment during the time period t under the scenario s; are respectively the charging and discharging powers of the energy storage device installed at the m-th heat storage type industrial load equipment during the time period t under the scenario s.

[0106] Exemplarily, the carbon emission cost of the heat storage type enterprise in the net revenue objective function obtained by the heat storage type enterprise can be expressed as:

[0107]

[0108] Among them, is the carbon emission cost of the heat storage type equipment m during the time period t under the scenario s; η M is the carbon emission factor per unit of electricity for the internal process of the heat storage type equipment; is the dynamic carbon emission factor per unit of electricity for power consumption, which is related to the proportion of fossil fuel power generation in the power grid where the heat storage type enterprise is located; η M,p is the carbon emission quota per unit of electricity for the heat storage type equipment.

[0109] In summary, the net revenue objective function obtained by the heat storage type enterprise can be expressed as:

[0110]

[0111] Among them, B M is the net revenue obtained by the electrolytic enterprise participating in demand response; N M is the number of heat storage type equipment in the industrial park microgrid. is the carbon emission cost of the heat storage type equipment m during the time period t under the scenario s; The revenue of the m-th heat storage equipment participating in demand response during period t in scenario s; A m,t,s The product profit loss caused by the actual standby power called by the m-th heat storage industrial equipment during period t; p s The probability of the occurrence of scenario s. T N The total number of periods in a day; N s The number of scenarios.

[0112] In summary, according to the net revenue B obtained by the electrolytic enterprises AL and the net revenue B obtained by the heat storage enterprises M , the objective function corresponding to the construction of the microgrid side model in the industrial park can be as follows:

[0113] max B AL +B M (31)

[0114] Among them, the constraint conditions corresponding to this objective function include formulas (5)-(19), and the objective of this objective function is to maximize the total net revenue of all electrolytic and heat storage industrial enterprises in the industrial park.

[0115] Exemplarily, the power balance constraint under the benchmark scenario can be expressed as:

[0116]

[0117] Among them, P t buy is the power purchased from the main grid by the distribution network during period t; P i,t is the output of the i-th gas turbine in the distribution network during period t; W j,t is the predicted power of the new energy power station j in the distribution network during period t; L d,t is the predicted load power of node d in the distribution network during period t; N g 、N W 、N D are the total numbers of gas turbines, new energy power stations and load nodes in the distribution network respectively. is the benchmark power of the m-th heat storage industrial load operation, P l,t is the power of the electrolytic equipment l during period t.

[0118] Exemplarily, the gas turbine output constraint under the benchmark scenario can be expressed as:

[0119] I i,t P i min ≤P i,t ≤I i,t P i max (33)

[0120] Among them, Pi min , P i max are the lower and upper limits of the output of the i-th gas turbine unit respectively. I i,t is the operating state of the i-th gas turbine unit at time period t. 0 indicates that the unit is in the shutdown state, and 1 indicates that the unit is in the operating state. P i,t is the output of the i-th gas turbine unit in the distribution network at time period t.

[0121] Exemplarily, the ramp rate constraint of the gas turbine unit under the reference scenario can be expressed as:

[0122] -DR i ≤P i,t -P i,t-1 ≤UR i (34)

[0123] where DR i , UR i are the upper and lower ramp rates of the thermal power unit i respectively; P i,t-1 is the output of the i-th gas turbine unit in the distribution network at time period t-1. P i,t is the output of the i-th gas turbine unit in the distribution network at time period t.

[0124] Exemplarily, the minimum start-stop time constraint of the gas turbine unit under the reference scenario can be expressed as:

[0125]

[0126] where are the continuous on and off times of the gas turbine unit i at time period t-1 respectively; are the minimum on and off durations of the gas turbine unit i respectively; I i,t-1 is the operating state of the i-th gas turbine unit at time period t-1.

[0127] Exemplarily, the reserve capacity constraint of the gas turbine unit under the reference scenario can be expressed as:

[0128]

[0129] where are the upper and lower reserve capacities reserved by the gas turbine unit i at time period t respectively. DR i , UR i are the upper and lower ramp rates of the thermal power unit i respectively; P i,t is the output of the i-th gas turbine unit in the distribution network at time period t. P i min , P i max are the lower and upper limits of the output of the i-th gas turbine unit respectively.

[0130] Exemplarily, the branch power flow constraint under the base scenario can be expressed as:

[0131]

[0132] where S P , S W , S L , S MG are the sensitivity matrices of the power of gas turbines, new energy power plants, node loads, and the total power of the industrial park microgrid with respect to the power flow of the distribution network lines respectively; P is the matrix composed of the powers of all gas turbines in the distribution network at all times; W is the matrix composed of the powers of all new energy power plants at all times; L is the matrix composed of the load powers of all nodes at all times; P AL , P M are the matrices composed of the powers consumed by all electrolytic and heat storage enterprises in the industrial park at all times respectively; is the matrix composed of the maximum values of the power flows of each line.

[0133] Exemplarily, the scenario energy balance constraint under the prediction error scenario can be expressed as:

[0134]

[0135] where are the up and down reserve capacities actually dispatched by gas turbine i at time t respectively; is the curtailment power value of new energy power plant j at time t under scenario s; is the actual load power value of node d at time t under scenario s. N g , N W , N D are the total numbers of gas turbines, new energy power plants, and load nodes in the distribution network respectively.

[0136] Exemplarily, the scenario unit ramp rate constraint under the prediction error scenario can be expressed as:

[0137]

[0138] where are the up and down reserve capacities actually dispatched by gas turbine i at time t - 1 respectively. DR i , UR i are the up and down ramp rates of thermal power unit i respectively.

[0139] Exemplarily, the scenario branch power flow constraint under the prediction error scenario can be expressed as:

[0140]

[0141] Among them, is the matrix composed of the reserve power actually called by all gas turbines in all time periods under scenario s; ΔW s is the matrix composed of the power prediction errors of all new energy power stations in all time periods under scenario s; is the matrix composed of the curtailed power of all new energy power stations in all time periods under scenario s; ΔL s is the matrix composed of the load power prediction errors of all nodes in all time periods under scenario s; is the matrix composed of the load shedding power of all nodes in all time periods under scenario s; ΔP AL,s and ΔP M,s are respectively the matrices composed of the reserve capacities actually called by all electrolytic and thermal energy storage enterprises in all time periods under scenario s; and are respectively the matrices composed of the powers of the energy storage devices installed at all electrolytic and thermal energy storage enterprises in all time periods under scenario s. S P 、S W 、S L 、S MG are respectively the sensitivity matrices of the gas turbine power, new energy power station power, node load power, and total power of the industrial park microgrid with respect to the power flow of the distribution network lines; is the matrix composed of the maximum values of the power flows of each line.

[0142] Exemplarily, the upper and lower reserve capacity call constraints of the unit under the prediction error scenario can be expressed as:

[0143]

[0144] Among them, are respectively the upper and lower reserve capacities reserved by gas turbine i in time period t. are respectively the upper and lower reserve capacities actually called by gas turbine i in time period t.

[0145] Exemplarily, the new energy curtailment power constraint under the prediction error scenario can be expressed as:

[0146]

[0147] Among them, is the curtailment power value of new energy power station j in time period t under scenario s; W j,t is the predicted power of new energy power station j in the distribution network in time period t; ΔW j,t,s represents the output power prediction error of the jth new energy power station in the sth typical scenario in time period t.

[0148] Exemplarily, the load shedding power constraint under the prediction error scenario can be expressed as:

[0149]

[0150] Among them, ΔL d,t,s represents the prediction error of the output power of the d-th load node in the s-th typical scenario at time period t; L d,t is the predicted load power of node d in the distribution network at time period t; is the actual load power value of node d at time period t under scenario s.

[0151] In summary, the objective function corresponding to the two-layer dispatching model can be expressed by the following formula:

[0152]

[0153] Among them, is the fuel cost per unit of power generation of gas turbine unit i; and are the compensation prices for the reserved upper and lower reserve capacities of gas turbine unit i respectively; and are the additional subsidy prices for the actually called upper and lower reserves of gas turbine unit i respectively; C load is the load shedding compensation price; C w is the penalty coefficient for abandoning new energy per unit of power generation. P i,t is the output of the i-th gas turbine unit in the distribution network at time period t; are the upper and lower reserve capacities reserved by gas turbine unit i at time period t respectively. is the abandoned power value of new energy power station j at time period t under scenario s; is the actual load power value of node d at time period t under scenario s. N g 、N W 、N D are the total numbers of gas turbine units, new energy power stations and load nodes in the distribution network respectively. are the upper and lower reserve capacities actually called by gas turbine unit i at time period t respectively; T N is the total number of time periods in a day; N s is the number of scenarios. p s is the probability of the occurrence of scenario s. B AL is the net income obtained by the electrolysis-type enterprise; B M is the net income obtained by the heat storage-type enterprise.

[0154] In summary, the double-layer scheduling model provided by the present invention is jointly composed of the objective function formula (44), the constraint conditions formulas (5)-(19) and (32)-(43). Specifically, the upper-layer model is an optimal scheduling model on the distribution network side, with the minimum total operating cost of the distribution network as the optimization objective. Its optimization results are the working states and powers of flexible resources such as the power purchased from the main grid, the power generation of new energy power stations, the start-stop states of gas turbines, the power of gas turbines, energy storage, and demand response, etc.

[0155] S103. Based on a preset solver, solve the double-layer scheduling model to obtain the target values of the grid power equipment for scheduling the grid power equipment.

[0156] Among them, the target values can be the parameters involved in the double-layer scheduling model and related constraint conditions, and can include at least one of the following: the net income obtained by electrolysis enterprises, the net income obtained by heat storage enterprises, the power purchased from the main grid by the distribution network, the output of gas turbines in the distribution network, the predicted power of new energy power stations in the distribution network, and the predicted load power of nodes in the distribution network. The power purchased from the main grid by the distribution network during a time period can be obtained by combining the double-layer scheduling model with the constraint condition formula (32).

[0157] Optionally, based on a preset solver, solve the double-layer scheduling model to obtain the target values of the grid power equipment for scheduling the grid power equipment, including: input the double-layer scheduling model and its corresponding constraint conditions into the preset solver, solve the double-layer scheduling model to obtain the target values of the grid power equipment; determine the scheduling plan for the grid power equipment according to the power purchased from the main grid and the output of gas turbines in the target values for scheduling the grid power equipment.

[0158] Correspondingly, after obtaining the target values of the grid power equipment, it further includes: determining the dynamic carbon emission factor according to the power purchased from the main grid and the output of gas turbines in the target values, in combination with the carbon emission intensity of the gas turbines and the average carbon emission factor per unit of electricity purchased from the main grid; determining the total load curve of the industrial park microgrid according to the dynamic carbon emission factor and the lower-layer model, and adjusting the double-layer scheduling model according to the total load curve and the upper-layer model to update the target values of the grid power equipment and realize the iteration between the upper and lower layer models.

[0159] Exemplarily, according to the power purchased from the main grid and the output of gas turbines in the target values, in combination with the carbon emission intensity of the gas turbines and the average carbon emission factor per unit of electricity purchased from the main grid, the dynamic carbon emission factor can be determined by the following formula:

[0160]

[0161] Among them, is the carbon emission intensity of the i-th gas turbine; η mainIt is the average carbon emission factor per unit of electricity purchased from the main grid externally. P t buy It is the power purchased from the main grid by the distribution network at time t; P i,t It is the output of the i-th gas turbine unit in the distribution network at time t; N g They are gas turbine units in the distribution network;

[0162] Exemplarily, the dynamic carbon emission factor (for example, the dynamic carbon emission factor can be calculated and determined once per hour) can be substituted into the lower-layer model, so as to optimize the total load curve of the industrial park microgrid under the condition of maximizing the benefits of enterprises in the park. Return this curve to the upper-layer model, then the corresponding dispatching plan can be continuously optimized and adjusted accordingly, and the dynamic carbon emission factor per hour is updated again, thus forming an iteration between the upper and lower-layer models.

[0163] It should be noted that the present invention takes into account that the carbon emission factor of the industrial park purchasing electricity from the distribution network changes dynamically with the proportion of the load demand electricity borne by fossil energy power generation in the distribution network. That is to say, the change of the carbon emission factor of purchasing electricity from the distribution network will affect the shape of the load power curve of each enterprise in the industrial park microgrid after participating in demand response; and the change of the power curve of the enterprises in the industrial park will, in turn, affect the proportion of the electricity quantity borne by fossil energy power generation in the distribution network, thereby changing the dynamic carbon emission factor of purchasing electricity from the distribution network per hour. That is to say, there is a coupling relationship between the optimal dispatching model of the industrial park microgrid and the optimal dispatching model of the distribution network it is connected to. Therefore, the present invention establishes a two-layer dispatching model including an upper-layer model and a lower-layer model to dynamically optimize parameters and dynamically adjust the carbon emission factor, so as to achieve the effect of energy conservation and carbon reduction on the basis of taking into account the operation economy and safety of the enterprises in the distribution network and the industrial park microgrid.

[0164] Optionally, before solving the two-layer dispatching model based on a preset solver, it further includes: based on the KKT conditions and the big M method, converting the two-layer dispatching model into a single-layer quadratic programming model; correspondingly, solving the two-layer dispatching model based on a preset solver includes: inputting the single-layer quadratic programming model into the preset solver to obtain the target values of the grid power equipment for dispatching the grid power equipment.

[0165] Optionally, the lower-layer model can be converted into the corresponding Kuhn-Tucker (KKT) conditions, and the KKT conditions are used as supplementary constraint conditions (in the form of partial derivatives) of the upper-layer model; according to the upper-layer model and the supplementary constraint conditions, the big M method is used to linearize the bilinear terms (that is, the terms obtained by multiplying the kkt multiplier and the optimization variable) in the upper-layer model to obtain a quadratic programming model.

[0166] It should be noted that considering that the two - layer scheduling model cannot be directly solved, the KKT conditions can be used to transform the two - layer model into a single - layer model. Since the above two - layer model is too complex to show the KKT transformation process, the two - layer scheduling model is abstractly expressed in a compact general form as follows:

[0167]

[0168] Among them, \(x\) and \(y\) are the decision variables of the upper and lower layers respectively; \(F(x,y)\) and \(f(x,y)\) are the objective functions of the upper and lower layers respectively; \(G\) a (x,y)\(\leq0\) and \(H\) b (x,y)\( = 0\) are the upper - layer inequality and equality constraint conditions respectively; \(g\) i (x,y)\(\leq0\) and \(h\) j (x,y)\( = 0\) are the lower - layer inequality and equality constraint conditions respectively.

[0169] Furthermore, the Lagrange multiplier method is used to transform the lower - layer optimization problem into its KKT conditions, and the augmented Lagrangian function model is constructed as follows:

[0170]

[0171] Among them, \(\lambda\) i , \(\mu\) j are the dual variables of the lower - layer constraint conditions; \(L(y,\lambda,\mu;x)\) is the augmented Lagrangian function corresponding to the lower - layer model.

[0172] Take the partial derivatives of the augmented Lagrangian function and set the partial derivatives equal to 0 to form a system of equations, and the KKT conditions can be obtained. By solving the KKT conditions, the relationship between \(x\) and \(y\) can be obtained:

[0173]

[0174] Combine the KKT conditions of the transformed lower - layer model with the constraints of the upper - layer model, so as to transform the two - layer model into a single - layer model:

[0175]

[0176] Solving the single - layer model formula (49) obtained by KKT transformation is equivalent to solving the two - layer scheduling model of the present invention.

[0177] It should be noted that since the single - layer model after the KKT condition transformation in the above steps contains a bilinear term of the product of the Lagrange multiplier and the decision variable, that is, the complementary slackness condition \(\lambda\) i g i (x,y)\( = 0\).

[0178] The general big - M method can be used to approximately transform this non - linear structure into the following linear constraints:

[0179]

[0180] where M big is a constant large enough; z i are several 0-1 auxiliary variables introduced by the big M method. λ i is the dual variable of the lower-layer constraint condition; g i (x, y) ≤ 0 is the lower-layer inequality constraint condition.

[0181] Through the above steps, the two-layer scheduling model can be transformed into a single-layer mixed-integer quadratic programming problem, so that it can be directly solved by calling commercial solvers such as Gurobi to obtain the target values of the grid power equipment, and an optimal scheduling plan for each power equipment and flexible resource in the distribution network and the industrial park microgrid can be generated to realize the scheduling of the grid power equipment.

[0182] The technical solution of the embodiment of the present invention responds to the scheduling request for the grid power equipment, clusters the historical data of the prediction error of the distribution network based on the preset density clustering algorithm to construct a prediction error scenario set according to the clustering result; determines the constraint conditions of the two-layer scheduling model according to the prediction error scenario set to construct a two-layer scheduling model; based on the preset solver, solves the two-layer scheduling model to obtain the target values of the grid power equipment for scheduling the grid power equipment. Through comprehensive analysis from both the distribution network side and the industrial park microgrid side, an accurate scheduling model can be established, accurate scheduling values can be determined, and optimal scheduling of the grid power equipment can be realized, so that the effect of energy conservation and carbon reduction can be achieved on the basis of taking into account the economic efficiency and safety of the operation of enterprises in the distribution network and the industrial park microgrid.

[0183] Embodiment 2

[0184] Figure 2 is the structural diagram of a scheduling device for grid power equipment provided in Embodiment 2 of the present invention; this embodiment is applicable to the situation of predicting the operation conditions of the distribution network and the industrial park microgrid based on the two-layer scheduling model to effectively schedule the grid power equipment. The scheduling device for grid power equipment provided in the embodiment of the present invention can execute the scheduling method for grid power equipment provided in any embodiment of the present invention, and has the corresponding function modules and beneficial effects of the execution method; the scheduling device for grid power equipment can be implemented in the form of hardware and / or software and is configured in an electronic device with the scheduling function of grid power equipment, such as the total control device of the distribution network and the workbench park microgrid, such as Figure 2 As shown, the scheduling device for grid power equipment specifically includes:

[0185] The clustering module 201 is configured to, in response to a scheduling request for grid power equipment, perform clustering processing on historical prediction error data of a distribution network based on a preset density clustering algorithm, so as to construct a prediction error scenario set according to the clustering result;

[0186] The construction module 202 is configured to determine constraint conditions of a two-layer scheduling model according to the prediction error scenario set, so as to construct the two-layer scheduling model; wherein, the two-layer scheduling model includes an upper-layer model and a lower-layer model, the upper-layer model is a distribution network side model, and the lower-layer model is an industrial park microgrid side model;

[0187] The scheduling module 203 is configured to solve the two-layer scheduling model based on a preset solver to obtain a target value of the grid power equipment, so as to schedule the grid power equipment.

[0188] The technical solution of the embodiment of the present invention, in response to a scheduling request for grid power equipment, performs clustering processing on historical prediction error data of a distribution network based on a preset density clustering algorithm, so as to construct a prediction error scenario set according to the clustering result; determines constraint conditions of a two-layer scheduling model according to the prediction error scenario set, so as to construct the two-layer scheduling model; solves the two-layer scheduling model based on a preset solver to obtain a target value of the grid power equipment, so as to schedule the grid power equipment. By comprehensively analyzing from both the distribution network side and the industrial park microgrid side, an accurate scheduling model can be established, an accurate scheduling value can be determined, and optimal scheduling of grid power equipment can be realized, so that the effect of energy conservation and carbon reduction can be achieved on the basis of taking into account the economic efficiency and safety of the operation of enterprises in the distribution network and the industrial park microgrid.

[0189] Further, the clustering module 201 is specifically configured to:

[0190] Perform clustering processing on historical prediction error data of a distribution network based on a preset density clustering algorithm, so as to divide the historical prediction error data into at least two categories;

[0191] For each divided category, based on a center clustering algorithm, determine a target clustering center from the target data points included in each category, and generate a prediction error scenario set according to the target clustering centers corresponding to each divided category.

[0192] Further, the constraint conditions of the two - layer scheduling model include the constraint conditions under the benchmark scenario and the constraint conditions under the prediction error scenario; the constraint conditions under the benchmark scenario include at least one of the following: power balance constraint, gas turbine output constraint, gas turbine ramping constraint, minimum start - stop time constraint of gas turbine, gas turbine reserve capacity constraint, and branch power flow constraint; the constraint conditions under the prediction error scenario include at least one of the following: scenario energy balance constraint, scenario unit ramping constraint, scenario branch power flow constraint, upper and lower reserve capacity call constraint of units, new - energy abandonment power constraint, load shedding power constraint, and enterprise constraints of electrolysis - type and heat - storage - type enterprises in the industrial park microgrid.

[0193] Further, the enterprise constraints of electrolysis - type and heat - storage - type enterprises in the industrial park microgrid include at least one of the following: electrolysis - type equipment power constraint, electrolysis - type equipment reserve capacity constraint, electrolysis - type equipment temperature constraint, daily output constraint of electrolysis - type enterprises, actual call constraint of electrolysis - type equipment reserve capacity, power - energy constraint corresponding to electrolysis - type equipment, daily regulation capacity constraint of heat - storage - type load, daily regulation times constraint of heat - storage - type industrial load, daily output constraint of heat - storage - type industrial load, and power - energy constraint corresponding to heat - storage - type equipment.

[0194] Further, the scheduling module 203 is specifically used for:

[0195] Input the two - layer scheduling model and its corresponding constraint conditions into a preset solver, solve the two - layer scheduling model, and obtain the target values of the grid power equipment;

[0196] Determine the scheduling scheme for the grid power equipment according to the purchased power and gas turbine output in the target values, so as to schedule the grid power equipment;

[0197] Correspondingly, the above - mentioned device is also used for:

[0198] Determine the dynamic carbon emission factor according to the purchased power and gas turbine output in the target values, combined with the carbon emission intensity of the gas turbine and the average carbon emission factor per unit power of the main grid's externally - purchased electricity;

[0199] Determine the total load curve of the industrial park microgrid according to the dynamic carbon emission factor and the lower - layer model, and adjust the two - layer scheduling model according to the total load curve and the upper - layer model, so as to update the target values of the grid power equipment and realize the iteration between the upper and lower layers of the model.

[0200] Further, the above - mentioned device is also used for:

[0201] Based on the Kuhn - Tucker (KKT) conditions and the law of large numbers, transform the two - layer scheduling model into a single - layer quadratic programming model;

[0202] Correspondingly, solving the two - layer scheduling model based on the preset solver includes:

[0203] Input the single-layer quadratic programming model into a preset solver to obtain the target values of the power grid equipment, so as to schedule the power grid equipment.

[0204] Embodiment III

[0205] Figure 3 It is a schematic structural diagram of the electronic device provided in Embodiment III of the present invention. Figure 3 It shows a schematic structural diagram of an electronic device 10 that can be used to implement the embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, personal digital processors, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0206] As Figure 3 shown, the electronic device 10 includes at least one processor 11, and a memory communicatively connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc. Among them, the memory stores a computer program executable by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.

[0207] Multiple components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disc, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0208] The processor 11 may be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the scheduling method of the grid power device.

[0209] In some embodiments, the scheduling method of the grid power device may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed onto the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the scheduling method of the grid power device described above may be executed. Alternatively, in other embodiments, the processor 11 may be configured to execute the scheduling method of the grid power device by any other suitable means (e.g., by means of firmware).

[0210] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs, the one or more computer programs being executable and / or interpretable on a programmable system including at least one programmable processor, the programmable processor being a special or general-purpose programmable processor, receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0211] The computer program for implementing the method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowchart and / or block diagram are implemented. The computer program can be executed entirely on the machine, partially on the machine, as a stand-alone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0212] In one embodiment, the embodiment of the present invention further includes a computer program product. The computer program product includes a computer program which, when executed by a processor, implements the scheduling method for grid power equipment according to any embodiment of the present invention.

[0213] It should be understood that various forms of the processes shown above can be used, steps can be reordered, added or deleted. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is imposed herein. The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for dispatching power equipment in a power grid, characterized in that: include: In response to a dispatch request for a power grid power device, clustering the prediction error historical data of the distribution network based on a preset density clustering algorithm to construct a prediction error scenario set according to the clustering results; According to the prediction error scenario set, the constraint conditions of the two-layer dispatch model are determined to construct the two-layer dispatch model; wherein the two-layer dispatch model includes an upper model and a lower model, the upper model is the distribution network side model, and the lower model is the industrial park microgrid side model; Based on the preset solver, the two-layer dispatch model is solved to obtain the target value of the power equipment of the power grid so as to dispatch the power equipment of the power grid; Among them, the objective function corresponding to the two-layer scheduling model is expressed by the following formula: in, is the fuel cost per unit of power generation of gas generator set i; and are the compensation prices for reserving upper and lower spare capacities for gas generator set i, respectively; and are the additional subsidy prices for the gas generator set i to be actually called up and down for standby respectively; C load is the load shedding compensation price; C w P is the penalty coefficient for abandoning the unit power generation of new energy; i,t The output of the i-th gas generator set in the distribution network during time period t; are the upper and lower spare capacities reserved for gas generator set i in time period t respectively; is the abandoned power value of the new energy station j in time period t under scenario s; is the actual load power value of node d in time period t under scenario s; N g 、N W 、N D are the total number of gas units, new energy stations and load nodes in the distribution network; are the upper and lower reserve capacities actually used by gas generator set i in period t; T N is the total number of time periods in a day; N s is the number of scenes; p s is the probability of scenario s occurring; B AL is the net income obtained by the electrolysis enterprise; B M is the net income obtained by the thermal storage enterprise; Among them, the objective function corresponding to the industrial park microgrid side model is: max B AL +B M .

2. The method according to claim 1, characterized in that Based on the preset density clustering algorithm, the prediction error historical data of the distribution network is clustered to construct a prediction error scenario set according to the clustering results, including: Based on a preset density clustering algorithm, clustering is performed on the prediction error historical data of the distribution network to divide the prediction error historical data into at least two categories; For each divided class, based on the center clustering algorithm, the target cluster center is determined from the target data points contained in each class, and a prediction error scenario set is generated according to the target cluster center corresponding to each divided class.

3. The method according to claim 1, characterized in that in, The constraints of the two-layer scheduling model include constraints under the baseline scenario and constraints under the prediction error scenario; the constraints under the baseline scenario include at least one of the following: power balance constraints, gas unit output constraints, gas unit climbing constraints, gas unit minimum start and stop time constraints, gas unit spare capacity constraints and branch flow constraints; the constraints under the prediction error scenario include at least one of the following: scenario energy balance constraints, scenario unit climbing constraints, scenario branch flow constraints, unit upper and lower spare capacity call constraints, new energy abandonment power constraints, load shedding power constraints and enterprise constraints of electrolysis and heat storage enterprises in industrial park microgrids.

4. The method according to claim 3, characterized in that in, The enterprise constraints of electrolysis and heat storage enterprises within the industrial park microgrid include at least one of the following: power constraints of electrolysis equipment, spare capacity constraints of electrolysis equipment, temperature constraints of electrolysis equipment, daily output constraints of electrolysis enterprises, actual call constraints of spare capacity of electrolysis equipment, power and energy constraints corresponding to electrolysis equipment, daily regulation capacity constraints of heat storage loads, daily regulation times constraints of heat storage industrial loads, daily output constraints of heat storage industrial loads, and power and energy constraints corresponding to heat storage equipment.

5. The method according to claim 1, characterized in that Based on the preset solver, the two-layer dispatch model is solved to obtain the target value of the power equipment of the power grid to dispatch the power equipment of the power grid, including: The two-layer dispatch model and its corresponding constraints are input into a preset solver, the two-layer dispatch model is solved, and the target value of the power equipment of the power grid is obtained; According to the purchased power and the output of the gas-fired units in the target value, a dispatching plan for the power equipment of the power grid is determined to dispatch the power equipment of the power grid; Correspondingly, after obtaining the target value of the power equipment of the power grid, it also includes: Determine the dynamic carbon emission factor based on the purchased power and gas unit output in the target value, combined with the carbon emission intensity of the gas unit and the average carbon emission factor per unit of electricity purchased from the main grid; According to the dynamic carbon emission factor and the lower-level model, the total load curve of the industrial park microgrid is determined, and according to the total load curve and the upper-level model, the two-layer scheduling model is adjusted to update the target value of the power equipment of the power grid and realize the iteration between the upper and lower-level models.

6. The method according to claim 1, characterized in that Based on the preset solver, before solving the two-level scheduling model, it also includes: Based on Kuhn-Tucker KKT conditions and the method of large numbers, the two-level scheduling model is transformed into a single-level quadratic programming model; Accordingly, based on the preset solver, the two-layer scheduling model is solved, including: The single-layer quadratic programming model is input into the preset solver to obtain the target value of the power equipment in the power grid so as to schedule the power equipment in the power grid.

7. A dispatching device for power grid power equipment, characterized in that: include: A clustering module, for responding to a dispatch request for a power grid power device, clustering the prediction error historical data of the distribution network based on a preset density clustering algorithm, so as to construct a prediction error scenario set according to the clustering results; A construction module is used to determine the constraint conditions of the two-layer scheduling model according to the prediction error scenario set to construct the two-layer scheduling model; wherein the two-layer scheduling model includes an upper model and a lower model, the upper model is a distribution network side model, and the lower model is an industrial park microgrid side model; A scheduling module, used to solve the double-layer scheduling model based on a preset solver to obtain the target value of the power equipment of the power grid so as to schedule the power equipment of the power grid; Among them, the objective function corresponding to the two-layer scheduling model is expressed by the following formula: in, is the fuel cost per unit of power generation of gas generator set i; and are the compensation prices for reserving upper and lower spare capacities for gas generator set i, respectively; and are the additional subsidy prices for the gas generator set i to be actually called up and down for standby respectively; C load is the load shedding compensation price; C w P is the penalty coefficient for abandoning the unit power generation of new energy; i,t The output of the i-th gas generator set in the distribution network during time period t; are the upper and lower spare capacities reserved for gas generator set i in time period t respectively; is the abandoned power value of the new energy station j in time period t under scenario s; is the actual load power value of node d in time period t under scenario s; N g 、N W 、N D are the total number of gas units, new energy stations and load nodes in the distribution network; are the upper and lower reserve capacities actually used by gas generator set i in period t; T N is the total number of time periods in a day; N s is the number of scenes; p s is the probability of scenario s occurring; B AL is the net income obtained by the electrolysis enterprise; B M is the net income obtained by the thermal storage enterprise; Among them, the objective function corresponding to the industrial park microgrid side model is: max B AL +B M .

8. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for dispatching power equipment in a power grid according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the scheduling method for power grid power equipment according to any one of claims 1 to 6 when executed.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the computer program implements the method for dispatching power equipment of a power grid according to any one of claims 1 to 6.

Images