Virtual power plant cluster resource replacement method, device, electronic device and medium

Abstract

The embodiments of the present disclosure disclose a method, device, electronic device and medium for resource replacement of a virtual power plant cluster. A specific implementation of the method includes: obtaining a distributed energy data set and a load data set of a virtual power plant cluster; performing feature data extraction on the distributed energy data set and the load data set to obtain a distributed energy feature vector set and a load feature vector set; performing clustering processing on the distributed energy feature vector set and the load feature vector set to obtain a virtual power plant data class cluster set; generating a resource replacement model set of the virtual power plant cluster; generating a virtual power plant objective function and a virtual power plant constraint function set based on the virtual power plant data class cluster set and the resource replacement model set; generating resource replacement information; and performing resource replacement processing on the virtual power plant cluster based on the resource replacement information. This implementation can improve the accuracy of the generated resource replacement information, reduce the consumption of power resources and improve the system stability of the virtual power plant.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of computer technology, and in particular to a virtual power plant cluster resource replacement method, device, electronic device, and medium. Background Art

[0002] At present, virtual power plants can aggregate and optimize different distributed resources through advanced information and communication technologies to achieve efficient resource utilization and intelligent management. They can also achieve resource aggregation management, promote local consumption of new energy, and ensure safe and economical operation of power grids. For the replacement of virtual power plant clusters, the usual method is to optimize resource replacement for each virtual power plant in the virtual power plant cluster to obtain a resource replacement optimization information set. Then, the resource replacement optimization information set is fused to obtain resource replacement information for the virtual power plant cluster, and resource replacement processing is performed on the virtual power plant cluster based on the resource replacement information.

[0003] However, it is found in practice that when the above method is used to perform resource replacement operations on a virtual power plant cluster, the following technical problems often occur: when optimizing resource replacement for a single virtual power plant and integrating the resource replacement optimization information set, only the resource replacement within a single virtual power plant is considered, and the resource replacement between virtual power plants is not considered. The accuracy of the generated resource replacement information is low, resulting in unreasonable resource replacement of the virtual power plant cluster, causing some virtual power plants to have a large amount of idle power resources and some virtual power plants to have insufficient resources, resulting in power lines with transmission capacity being unable to carry too much electricity, high power line losses, and low stability of virtual power plants.

[0004] In the process of adopting technical solutions to solve the above technical problem 1, the following technical problem 2 is often accompanied: since the load data set contains a large amount of redundant and noisy data, it is easy to cause inaccurate prediction of load demand, and the accuracy of the virtual power plant objective function and virtual constraint condition function set of the virtual power plant cluster is low, which makes the resource allocation of the virtual power plant in the virtual power plant cluster unbalanced, resulting in low system stability of the virtual power plant cluster and high damage rate of the unit equipment included. For the above technical problem 2, the conventional solution is generally: using the empirical mode decomposition algorithm to perform data mode decomposition on the load data set to obtain the decomposed load data set. Then, the decomposed load data set is input into the autoregressive moving average model to obtain the load demand. However, the above conventional solution still has the following problems: due to the problem of mode aliasing when using the empirical mode decomposition algorithm for data decomposition, it is easy to cause incomplete decomposed data and the generation of noise data, and the autoregressive moving average model only predicts by focusing on the decomposed load data, and the feature information of the decomposed load model is not fully extracted, resulting in inaccurate prediction of load demand, resulting in low system stability of the virtual power plant cluster and high damage rate of the unit equipment included.

[0005] In the process of adopting technical solutions to solve the above-mentioned technical problem one, the following technical problem three is often accompanied: due to inaccurate solutions to the virtual power plant objective function and the virtual power plant constraint condition function set, the resource allocation in the virtual power plant cluster is unbalanced, resulting in waste of power resources generated by the virtual power plant cluster and low system stability of the virtual power plant cluster. For the above-mentioned technical problem three, the conventional solution is generally: using the particle swarm optimization algorithm to solve the virtual power plant objective function and the virtual power plant constraint condition function set to obtain resource replacement information. However, the above-mentioned conventional solution still has the following problems: due to random initialization in the particle swarm optimization algorithm, it is easy to cause uneven distribution of the initial position of the particle swarm, increase resource waste and prolong computing resources, and the particle swarm optimization algorithm is easy to fall into the local optimal solution, which makes the accuracy of the generated resource replacement information low, the virtual power plant cluster resource scheduling is not accurate, resulting in unreasonable resource allocation of the virtual power plant cluster, waste of generated power resources and low system stability of the virtual power plant cluster.

[0006] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure concept and therefore it may contain information that does not form the prior art that is already known in this country to a person of ordinary skill in the art. Summary of the invention

[0007] The content of this disclosure is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this disclosure is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.

[0008] Some embodiments of the present disclosure propose a virtual power plant cluster resource replacement method, device, electronic device and medium to solve one or more of the technical problems mentioned in the above background technology section.

[0009] In a first aspect, some embodiments of the present disclosure provide a method for resource replacement of a virtual power plant cluster, comprising: obtaining a distributed energy data set and a load data set of a virtual power plant cluster, wherein the distributed energy data set includes at least one of the following: distributed power source data, a storage system, a controllable load, electric vehicle charging data, and power equipment operating status information; performing feature data extraction on the distributed energy data set and the load data set to obtain a distributed energy feature vector set and a load feature vector set; performing clustering processing on the distributed energy feature vector set and the load feature vector set to obtain a virtual power plant data class cluster set; generating a resource replacement model for each virtual power plant in the virtual power plant cluster to obtain a resource replacement model set; generating a virtual power plant objective function and a virtual power plant constraint function set of the virtual power plant cluster based on the virtual power plant data class cluster set and the resource replacement model set; generating resource replacement information for the virtual power plant cluster based on the virtual power plant objective function and the virtual power plant constraint function set; and performing resource replacement processing on the virtual power plant cluster based on the resource replacement information.

[0010] In a second aspect, some embodiments of the present disclosure provide a virtual power plant cluster resource replacement device, including: an acquisition unit, configured to acquire a distributed energy data set and a load data set of a virtual power plant cluster, wherein the distributed energy data set includes at least one of the following: distributed power source data, a storage system, a controllable load, an electric vehicle charging data, and power equipment operating status information; a feature data extraction unit, configured to perform feature data extraction on the distributed energy data set and the load data set to obtain a distributed energy feature vector set and a load feature vector set; a clustering unit, configured to perform clustering on the distributed energy feature vector set and the load feature vector set Clustering processing is performed to obtain a virtual power plant data cluster set; the first generation unit is configured to generate a resource substitution model for each virtual power plant in the above-mentioned virtual power plant cluster to obtain a resource substitution model set; the second generation unit is configured to generate a virtual power plant objective function and a virtual power plant constraint function set of the above-mentioned virtual power plant cluster based on the above-mentioned virtual power plant data cluster set and the above-mentioned resource substitution model set; the third generation unit is configured to generate resource substitution information for the above-mentioned virtual power plant cluster based on the above-mentioned virtual power plant objective function and the above-mentioned virtual power plant constraint function set; the resource substitution unit is configured to perform resource substitution processing on the above-mentioned virtual power plant cluster based on the above-mentioned resource substitution information.

[0011] In a third aspect, some embodiments of the present disclosure provide an electronic device comprising: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner in the first aspect.

[0012] In a fourth aspect, some embodiments of the present disclosure provide a computer-readable medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method described in any implementation manner in the first aspect is implemented.

[0013] The above-mentioned embodiments of the present disclosure have the following beneficial effects: the virtual power plant cluster resource replacement method of some embodiments of the present disclosure can improve the accuracy of the generated resource replacement information, reduce the consumption of power resources and improve the system stability of the virtual power plant. Specifically, the reason why the relevant power lines with transmission capacity cannot carry too much electric energy, the power line loss is high, and the stability of the power grid is low is that: the resource replacement optimization of a single virtual power plant is carried out, and the resource replacement optimization information set is integrated, only the resource replacement within a single virtual power plant is considered, and the resource replacement between virtual power plants is not considered. The accuracy of the generated resource replacement information is low, resulting in unreasonable resource replacement of the virtual power plant cluster, so that some virtual power plant resources have a large amount of idle power resources, and some virtual power plants have insufficient resources, resulting in the power lines with transmission capacity cannot carry too much electric energy, the power line loss is high, and the stability of the virtual power plant is low. Based on this, the virtual power plant cluster resource replacement method of some embodiments of the present disclosure can first obtain the distributed energy data set and load data set of the virtual power plant cluster, wherein the above-mentioned distributed energy data set includes at least one of the following: distributed power supply data, storage system, controllable load, electric vehicle charging data, and power equipment operation status information. Here, the distributed energy data set and the load data set are convenient for subsequent clustering processing. Secondly, feature data extraction is performed on the above-mentioned distributed energy data set and the above-mentioned load data set to obtain a distributed energy feature vector set and a load feature vector set. Here, representative data can be extracted from the distributed energy data set and the load data set to reduce redundant data and data volume. Again, the above-mentioned distributed energy feature vector set and the above-mentioned load feature vector set are clustered to obtain a virtual power plant data class cluster set. Here, the differences and correlations between the virtual power plants included in the virtual power plant cluster can be determined, which is convenient for the subsequent resource replacement processing of the virtual power plant cluster. Next, a resource replacement model for each virtual power plant in the above-mentioned virtual power plant cluster is generated to obtain a resource replacement model set. Here, the resource replacement performance of the virtual power plant can be more accurately grasped, which is convenient for the subsequent resource replacement processing of the virtual power plant cluster. Subsequently, according to the above-mentioned virtual power plant data class cluster set and the above-mentioned resource replacement model set, the virtual power plant objective function and virtual power plant constraint function set of the above-mentioned virtual power plant cluster are generated. Here, more accurate virtual power plant objective functions and virtual power plant constraint function sets between and within virtual power plants can be obtained, so as to generate resource replacement information later. Then, based on the above virtual power plant objective function and the above virtual power plant constraint function set, resource replacement information for the above virtual power plant cluster is generated. Here, solving the virtual power plant objective function and the virtual power plant constraint function set can obtain resource replacement information with higher accuracy, reducing the waste of power resources within and between virtual power plants. Finally, based on the above resource replacement information, the above virtual power plant cluster is subjected to resource replacement processing.Here, the rationality of resource replacement of the virtual power plant cluster can be improved, the loss of power lines and the system stability of the virtual power plant cluster can be reduced. It can be concluded that the virtual power plant cluster resource replacement method can improve the accuracy of the generated resource replacement information, reduce the consumption of power resources and improve the system stability of the virtual power plant by considering the impact of resource replacement within and between virtual power plants. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that components and elements are not necessarily drawn to scale.

[0015] Figure 1 is a flow chart of some embodiments of a virtual power plant cluster resource replacement method according to the present disclosure;

[0016] Figure 2 It is a structural schematic diagram of some embodiments of a virtual power plant cluster resource replacement device according to the present disclosure;

[0017] Figure 3 It is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0018] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0019] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present disclosure can be combined with each other.

[0020] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0021] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0022] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0023] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0024] Figure 1 The flowchart 100 of some embodiments of the virtual power plant cluster resource replacement method according to the present disclosure is shown. The virtual power plant cluster resource replacement method includes the following steps:

[0025] Step 101, obtaining a distributed energy data set and a load data set of a virtual power plant cluster.

[0026] In some embodiments, the execution subject (e.g., electronic device) of the above-mentioned virtual power plant cluster resource replacement method can obtain the distributed energy data set and load data set of the virtual power plant cluster through a wired connection method or a wireless connection method. Among them, the distributed energy data in the above-mentioned distributed energy data set can be various energy data included in the virtual power plant. The above-mentioned various energy data may include but are not limited to at least one of the following: distributed power supply data, storage system, controllable load, electric vehicle charging data, and power equipment operation status information. The above-mentioned controllable load may include power consumption data of industrial power load and civil power load. The above-mentioned resource unit equipment distribution information may be the power generation information and equipment status information of power generation equipment of various energy data. The load data in the above-mentioned load data set may be the power load demand information in a certain area covered by the virtual power plant cluster. The above-mentioned power load demand information may be the information of the electric power consumed by the power equipment in any time period. The above-mentioned load data set may include: power consumption spatiotemporal distribution data, power consumption value elasticity data, power demand and power generation data set of virtual power plants under various weather environments. The above-mentioned power consumption spatiotemporal distribution data can characterize the time characteristic information and spatial characteristic information of the power consumption of users in different geographical locations in different time periods. The above electricity consumption value elasticity data can characterize the trend of user electricity demand as the power generation value (electricity price) changes.

[0027] Step 102: extract feature data from the distributed energy data set and the load data set to obtain a distributed energy feature vector set and a load feature vector set.

[0028] In some embodiments, the above-mentioned execution subject may perform feature data extraction on the above-mentioned distributed energy data set and the above-mentioned load data set to obtain a distributed energy feature vector set and a load feature vector set. Among them, the above-mentioned distributed energy feature vector set may characterize the feature information of the distributed energy data set. The distributed energy feature vectors included in the above-mentioned distributed energy feature vector set may be a set of linearly unrelated feature vectors. The above-mentioned load feature vector set may characterize the feature information set of the load data set. The load feature vectors included in the above-mentioned load feature vector set may be a set of linearly unrelated feature vectors. In practice, the above-mentioned execution subject may use principal component analysis to perform feature data extraction on the above-mentioned distributed energy data set and the above-mentioned load data set to obtain a distributed energy feature vector set and a load feature vector set.

[0029] Step 103, clustering the distributed energy feature vector set and the load feature vector set to obtain a virtual power plant data cluster set.

[0030] In some embodiments, the execution subject may perform clustering processing on the distributed energy feature vector set and the load feature vector set to obtain a virtual power plant data cluster set. The virtual power plant data cluster in the virtual power plant data cluster set may be a set consisting of multiple distributed energy feature vectors and multiple load feature vectors with a certain similarity. The certain similarity may be a cosine similarity greater than or equal to 0.8 determined using cosine similarity.

[0031] In some optional implementations of some embodiments, clustering the distributed energy feature vector set and the load feature vector set to obtain a virtual power plant data cluster set may include the following steps:

[0032] The first step is to perform data normalization processing on the above-mentioned distributed energy feature vector set and the above-mentioned load feature vector set to obtain a standardized feature vector set. In practice, the above-mentioned execution entity can use the Z-Score (Z-Score Normalization) normalization algorithm to perform data normalization processing on the above-mentioned distributed energy feature vector set and the above-mentioned load feature vector set to obtain a standardized feature vector set.

[0033] The second step is to determine the target cluster value corresponding to the standardized feature vector set. The target cluster value may be the number of clusters formed after clustering the standardized feature vector set. In practice, the execution subject may use the Elbow Method to determine the target cluster value corresponding to the standardized feature vector set.

[0034] The third step is to select the target cluster numerical standardized feature vectors from the standardized feature vector set as the initial cluster center feature vector set. The screening can be performed using a roulette wheel selection algorithm.

[0035] The fourth step is to perform the following cluster center determination steps based on the initial cluster center feature vector set:

[0036] Sub-step 1, determining the characteristic distance value of each residual standardized feature vector in the residual standardized feature vector set and each initial cluster center feature vector in the initial cluster center feature vector set, and obtaining a characteristic distance value set. The residual standardized feature vector set may be a feature vector set obtained by removing the initial cluster center feature vector set from the standardized feature vector set. The characteristic distance value may be a Euclidean distance.

[0037] Sub-step 2, according to the feature distance value set, determine the initial cluster center feature vector corresponding to each residual standardized feature vector in the residual standardized feature vector set, and obtain the initial virtual power plant data cluster set. The initial virtual power plant data cluster in the initial virtual power plant data cluster set can be a cluster formed by taking the initial cluster center feature vector as the cluster center and each residual standardized feature vector with the smallest distance value from the initial cluster center feature vector.

[0038] As an example, the execution subject may first perform the following determination step for each remaining standardized feature vector in the remaining standardized feature vector set: filter out the respective feature distance values ​​from the remaining standardized feature vector set to each initial cluster center feature vector in the initial cluster center feature vector set from the feature distance value group set as the target feature distance value set. Secondly, filter out the target feature distance with the smallest value from the target feature distance value set as the filtered feature distance value. Then, determine the initial cluster center feature vector corresponding to the filtered feature distance value as the cluster center feature vector of the remaining standardized feature vector.

[0039] Sub-step 3, determining the mean of the feature vectors of each initial virtual power plant data cluster in the initial virtual power plant data cluster set as the updated cluster center feature vector, and obtaining the updated cluster center feature vector set.

[0040] Sub-step 4, in response to determining that each updated cluster center feature vector in the updated cluster center feature vector set is the same as the initial cluster center feature vector corresponding to the updated cluster center feature vector in the initial cluster center feature vector set, determines the initial virtual power plant data class cluster set as the virtual power plant data class cluster set.

[0041] In the fifth step, in response to determining that each updated cluster center feature vector in the updated cluster center feature vector set is different from the initial cluster center feature vector corresponding to the updated cluster center feature vector in the initial cluster center feature vector set, the updated cluster center feature vector set is determined as the initial cluster center feature vector set to perform the above cluster center determination step again.

[0042] In some optional implementations of some embodiments, the above-mentioned determination of the target cluster value corresponding to the above-mentioned standardized feature vector set may include the following steps:

[0043] The first step is to determine the arithmetic square root of the number of eigenvectors of the standardized eigenvectors included in the above-mentioned standardized eigenvector set as the termination cluster value.

[0044] The second step is to set the initial cluster value of the target cluster value as a first preset threshold value. The first preset threshold value may be a preset value. For example, the first preset threshold value may be 1.

[0045] The third step is to perform the following cluster average evaluation coefficient screening steps based on the initial cluster values:

[0046] Sub-step 1: clustering the standardized feature vector set to obtain an initial cluster value feature vector clusters. The clustering process may be clustering using a k-means algorithm.

[0047] Sub-step 2, for each feature vector cluster in the initial cluster value feature vector clusters, perform the following clustering evaluation coefficient generation steps:

[0048] The first sub-step is to determine the average intra-cluster distance value of the standardized feature vector set included in each feature vector cluster in the initial cluster value feature vector clusters and the feature vector corresponding to the cluster center of each feature vector cluster as the average intra-cluster distance value and the central feature vector, and obtain the average intra-cluster distance value set and the central feature vector set. The above-mentioned average intra-cluster distance value can be the average value of the distance value set of each standardized feature vector included in the above-mentioned feature vector cluster from the standardized feature vector located at the cluster center.

[0049] The second sub-step is to determine the inter-cluster distance values ​​of the central feature vector corresponding to the feature vector cluster and each remaining central feature vector in the remaining central feature vector set to obtain an inter-cluster distance value set. The remaining central feature vector set may be a feature vector set obtained by removing the central feature vector from the central feature vector set corresponding to the initial cluster value feature vector clusters.

[0050] The third sub-step is to screen the residual central feature vector set according to the inter-cluster distance value set to obtain the target residual central feature vector, wherein the target residual central feature vector may be the residual central feature vector corresponding to the minimum inter-cluster distance value in the residual central feature vector set.

[0051] As an example, the execution subject may determine the smallest inter-cluster distance value in the inter-cluster distance value set as the target inter-cluster distance value, and then determine the remaining central feature vector corresponding to the target inter-cluster distance value as the target remaining central feature vector.

[0052] The fourth sub-step is to generate a clustering evaluation coefficient according to the average distance value within the cluster and the average value within the cluster corresponding to the target residual center feature vector. The clustering evaluation coefficient can be a coefficient that represents the clustering effect by the ratio of the clustering degree between the standardized feature vector sets within the feature vector cluster and the degree of dispersion between the feature vector cluster sets. The larger the clustering evaluation coefficient is, the better the clustering effect is.

[0053] As an example, the execution subject may first determine the difference between the intra-cluster average distance value and the intra-cluster average value corresponding to the target residual center feature vector to obtain the intra-cluster difference value. Then, determine the larger distance value between the intra-cluster average distance value and the intra-cluster average value corresponding to the target residual center feature vector as the target intra-cluster average distance value. Finally, determine the ratio of the intra-cluster difference value to the target intra-cluster average distance value as the clustering evaluation coefficient.

[0054] Sub-step 3: Determine the average value of each clustering evaluation coefficient obtained as the clustering average evaluation coefficient.

[0055] Sub-step 4: adding the cluster average evaluation coefficient to a preset cluster average evaluation coefficient list to obtain a cluster average evaluation coefficient list. The preset cluster average evaluation coefficient list may be a pre-set storage list for storing cluster average evaluation coefficients.

[0056] Sub-step 5: The sum of the initial cluster value and the second preset threshold is determined as the updated cluster value. The second preset threshold may be a preset value. For example, the second preset threshold may be 1.

[0057] Sub-step 6, in response to determining that the updated cluster value is greater than or equal to the above-mentioned termination cluster value, a cluster average evaluation coefficient that meets the preset condition is selected from the obtained cluster average evaluation coefficient list as the target cluster average evaluation coefficient, and the initial cluster value corresponding to the above-mentioned target cluster average evaluation coefficient is determined as the target cluster value. The above-mentioned preset condition can be the condition of the cluster average evaluation coefficient with the smallest value after selection in the above-mentioned cluster average evaluation coefficient list.

[0058] In the fourth step, in response to determining that the updated cluster value is less than the above-mentioned termination cluster value, the updated cluster value is determined as the initial cluster value to perform the above-mentioned cluster average evaluation coefficient screening step again.

[0059] Step 104 , generating a resource replacement model for each virtual power plant in the virtual power plant cluster to obtain a resource replacement model set.

[0060] In some embodiments, the execution entity may generate a resource replacement model for each virtual power plant in the virtual power plant cluster to obtain a resource replacement model set. The resource replacement model may be a factor set for evaluating the resource replacement performance of the virtual power plant. The resource replacement performance may include: seasonal characteristics of distributed energy sources included in the virtual power plant and distributed resource regulation performance.

[0061] In some optional implementations of some embodiments, the step of generating a resource substitution model for each virtual power plant in the virtual power plant cluster to obtain a resource substitution model set may include the following steps:

[0062] For each virtual power plant in the above virtual power plant cluster, the following weighted summation steps are performed:

[0063] Sub-step 1, in response to receiving resource scheduling information, determining the temperature control load power of the virtual power plant. The resource scheduling information may be the received resource scheduling information. The temperature control load power may represent the electric power that the virtual power plant needs to consume. For example, the temperature control load power function may be the electric power that the air conditioning cluster refrigeration system needs to consume when reaching a steady state. The temperature control load power may be

[0064]

[0065] Among them, P ac (t) represents the temperature control load power in the time period t. α represents the efficiency ratio of the air conditioner. β represents the indoor equivalent thermal resistance. T out Indicates the outdoor temperature. in represents the indoor temperature. E() represents the data expectation. M represents the number of air conditioners included in the air conditioner cluster.

[0066] Sub-step 2, according to the preset environmental threshold, determine the first temperature control load power and the second temperature control load power of the temperature control load power. The first temperature control load power may be the maximum value of the temperature control load power function. The second temperature control load power may be the minimum value of the temperature control load power function. The preset environmental threshold may be a preset indoor temperature threshold and a temperature margin. The first temperature control load power may be

[0067]

[0068] Among them, P ac，max (t) represents the first temperature control load power. set Indicates the preset indoor temperature threshold. Δ Represents the temperature margin, that is, the temperature change.

[0069]

[0070] Among them, P ac，min (t) represents the second temperature control load power.

[0071] As an example, the above-mentioned execution entity can input the above-mentioned preset environmental threshold into the temperature-controlled load power to obtain the first temperature-controlled load power and the second temperature-controlled load power.

[0072] Sub-step 3, determining the resource replacement response direction and resource replacement response depth of the virtual power plant according to the temperature control load power, the first temperature control load power and the second temperature control load power. The resource replacement response direction may characterize the adjustment of the output electric power by the virtual power plant after receiving the resource scheduling information. The resource replacement response direction includes: upward and downward. Upward characterizes the increase in output electric power after receiving the resource scheduling information. Downward characterizes the reduction in output electric power after receiving the resource scheduling information. The resource replacement response depth may characterize the maximum offset of downward or upward movement. The resource replacement response depth may include: downward response depth and upward response depth. The downward response depth may be expressed as:

[0073]

[0074] in, Indicates the downward response depth.

[0075]

[0076] in, Indicates the downward response depth.

[0077] As an example, the execution subject may first determine that the resource replacement response direction of the virtual power plant is an upward direction in response to determining that the preset indoor temperature threshold is less than the indoor temperature. In response to determining that the preset indoor temperature threshold is greater than the indoor temperature, the resource replacement response direction of the virtual power plant is determined to be a downward direction. Secondly, the ratio of the temperature-controlled load power to the second temperature-controlled load power is determined as the upward response depth of the resource replacement response depth. Finally, the ratio of the first temperature-controlled load power to the temperature-controlled load power is determined as the downward response depth of the resource replacement response depth.

[0078] Sub-step 4, determining the resource replacement response time, resource replacement response duration, and resource replacement response rate of the virtual power plant. Among them, the resource replacement response time can represent the shortest time required for the virtual power plant to complete the state required by the resource scheduling information after receiving the resource scheduling information. The resource replacement response duration can represent the longest time that the virtual power plant maintains the state required by the resource scheduling information after reaching the state required by the resource scheduling information. The resource replacement response rate can represent the speed at which the virtual power plant changes from the operating state before receiving the resource scheduling information to the operating state required by the resource scheduling information. The resource replacement response rate can be expressed as:

[0079]

[0080]

[0081] Among them, R ac (t) represents the resource replacement response rate. Indicates the lower limit of the resource replacement response rate. Indicates the upper limit of resource replacement response rate. ΔT out Indicates the change in outdoor temperature during a period of time. Indicates the maximum value of the upward change in indoor temperature. Indicates the maximum downward change in indoor temperature.

[0082] Sub-step 5, determining the resource replacement response direction, the resource replacement response depth, the resource replacement response time, the resource replacement response duration and the resource replacement response rate as the resource replacement model of the virtual power plant.

[0083] Step 105 , generating a virtual power plant objective function and a virtual power plant constraint function set of the virtual power plant cluster according to the virtual power plant data class cluster set and the resource replacement model set.

[0084] In some embodiments, the execution subject may generate the virtual power plant objective function and the virtual power plant constraint function set of the virtual power plant cluster according to the virtual power plant data class cluster set and the resource replacement model set. Among them, the virtual power plant objective function may be a function that characterizes the differential resource characteristics within each virtual power plant and between virtual power plants included in the virtual power plant cluster and the dynamic replacement of resources of multiple virtual power plants from the perspective of the balance of supply and demand of electricity consumption. The virtual power plant objective function may be a resource replacement function that takes the virtual power plant data class cluster set and the resource replacement model as known data and takes into account the balance of supply and demand and economic benefits of the virtual power plant. The virtual power plant constraint function in the virtual power plant constraint function set may be a function that takes into account the safety and cost constraints of the virtual power plant cluster in the form of a function. The virtual power plant constraint function set may include but is not limited to at least one of the following: a function that characterizes the flow balance constraint of the virtual power plant cluster, a function that characterizes the load constraint of the virtual power plant cluster, and a function that characterizes the voltage constraint of the virtual power plant.

[0085] In some optional implementations of some embodiments, the generating of the virtual power plant objective function and the virtual power plant constraint function set of the virtual power plant cluster according to the virtual power plant data class cluster set and the resource substitution model set may include the following steps:

[0086] The first step is to obtain the power acquisition value, load power and power consumption value of each virtual power plant data cluster in the above virtual power plant data cluster set, and obtain the power acquisition value set, load power set and power consumption value set. Among them, the above power acquisition value can be the unit power value benefit value of each virtual power plant in the virtual power plant cluster. The above load power can be the unit time output power of each virtual power plant in the virtual power plant cluster. The above power consumption value can be the basic power generation value (power generation cost value) per unit time of each virtual power plant in the above virtual power plant cluster.

[0087] The second step is to determine the power acquisition value function of the virtual power plant cluster by adding the product of each power acquisition value in the power acquisition value set and the load power in the load power set corresponding to the power acquisition value. The power acquisition value function can be expressed as:

[0088]

[0089] Where W1 represents the power acquisition value function. T represents the time period. N represents the number of virtual power plants included in the virtual power plant cluster. i (t) represents the power acquisition value of virtual power plant i. i (t) represents the load power of virtual power plant i.

[0090] The third step is to determine the power consumption value function of the virtual power plant cluster by adding the product of each power consumption value in the power consumption value set and the load power in the load power set corresponding to the power consumption value. The power consumption value function can be expressed as:

[0091]

[0092] Where W2 represents the electricity consumption value function. i (t) represents the value of electricity consumption.

[0093] In the fourth step, in response to determining that there is at least one resource replacement response depth less than a preset resource replacement response depth threshold in the resource replacement response depth set included in the resource replacement model set, the at least one resource replacement response depth is determined as a target resource replacement response depth set. The preset resource replacement response depth threshold may be a maximum value of a pre-determined resource replacement response depth.

[0094] The fifth step is to determine the replacement range penalty function of each virtual power plant corresponding to the target resource replacement response depth set. The replacement range penalty function can be expressed as:

[0095]

[0096] Where W3 represents the replacement range penalty function. N1 represents the number of virtual power plants corresponding to the target resource replacement response depth set. penaltyl，i (t) represents the deviation assessment item of the resource replacement response depth of virtual power plant i.

[0097] In step 6, in response to determining that there is at least one resource replacement response rate greater than a preset resource replacement response rate threshold in the resource replacement response rate set included in the resource replacement model set, at least one resource replacement response rate is determined as a target resource replacement response rate set. The preset resource replacement response rate threshold may be a maximum value of a preset resource replacement response rate.

[0098] The seventh step is to determine the response rate penalty function of each virtual power plant corresponding to the target resource replacement response rate set. The response rate penalty function can be expressed as:

[0099]

[0100] Where W4 represents the response rate penalty function. N2 represents the number of virtual power plants corresponding to the above target resource replacement response rate. penalty2，i (t) represents the deviation assessment item of the resource replacement response rate of virtual power plant i.

[0101] Step 8: Subtract the power acquisition value function, the power consumption value function, the replacement range penalty function and the response time penalty function to obtain the virtual power plant objective function. The virtual power plant objective function can be expressed as:

[0102]

[0103] In some optional implementations of some embodiments, the step of generating the virtual power plant objective function and the virtual power plant constraint function set of the virtual power plant cluster according to the virtual power plant data class cluster set and the resource substitution model set may further include the following steps:

[0104] The first step is to perform load forecasting on the load data set to obtain the load demand for the forecast time period. The load demand for the forecast time period may be the required power consumption in the area covered by the virtual power plant within a period of time. The load forecast may be a forecast using a load forecasting model. The load forecasting model may be a convolutional neural network model.

[0105] The second step is to determine the cumulative sum of the above load power sets as the target load power of the above virtual power plant cluster. The above target load power can be expressed as

[0106] The third step is to generate the power supply and demand balance constraint function of the virtual power plant cluster according to the difference between the target load power and the load demand in the forecast time period and the preset grid deviation threshold. The preset grid deviation threshold can be the maximum value of the preset grid deviation value. The power supply and demand balance constraint function can be expressed as:

[0107]

[0108] in, represents the load demand in the forecast period. D(t) represents the power demand per unit time. ε represents the preset grid deviation threshold. Indicates the target load power.

[0109] As an example, the execution entity may determine a constraint function in which the difference between the target load power set and the load demand in the forecast time period is greater than or equal to the negative of a preset grid deviation threshold and less than or equal to the preset grid deviation threshold as a power supply and demand balance constraint condition function.

[0110] The fourth step is to generate the equipment load constraint condition function of the virtual power plant cluster according to each load power in the load power set, the first preset load power threshold and the second preset load power threshold. The first preset load power threshold may be the preset maximum output power of each virtual power plant in the virtual power plant cluster. The second preset load power threshold may be the preset minimum output power of each virtual power plant in the virtual power plant cluster. The equipment load constraint condition function may be a constraint function in which the load power is greater than or equal to the second preset load power threshold and less than or equal to the first preset load power threshold.

[0111] As an example, the execution entity may determine a constraint function in which the load power is greater than or equal to the second preset load power threshold and less than or equal to the first preset load power threshold as the equipment load constraint condition function.

[0112] The fifth step is to determine the equipment start-stop constraint function of the cluster unit equipment set of the virtual power plant cluster. The equipment start-stop constraint function can represent the function of the startup delay and shutdown delay of the virtual power plant cluster. The equipment start-stop constraint function can be expressed as:

[0113]

[0114] Among them, t star.i represents the startup time of virtual power plant i. stpp.i represents the shutdown time of virtual power plant i.

[0115] Step 6: According to the ramp rate corresponding to each load power in the load power set and the resource replacement response rate set corresponding to the resource replacement model set, determine the response rate constraint function of the virtual power plant cluster. The ramp rate can represent the maximum rate of change of the output power of the distributed energy resources included in the virtual power plant within a unit time. The response rate constraint function can be expressed as:

[0116]

[0117] Among them, P i (t+1)-P i (t) represents the climbing rate. i (t+1) represents the load power of virtual power plant i at time t+1. up，i R represents the maximum value of the upward resource replacement response rate of virtual power plant i. down，i Represents the maximum value of the downward resource replacement response rate of virtual power plant i.

[0118] The seventh step is to determine the above-mentioned power supply and demand balance constraint condition function, the above-mentioned equipment load constraint condition function, the above-mentioned equipment start and stop constraint condition function and the above-mentioned response rate constraint condition function as the virtual power plant constraint function set of the above-mentioned virtual power plant cluster.

[0119] In some optional implementations of some embodiments, performing load forecasting on the load data set to obtain load demand in a forecast time period may include the following steps:

[0120] The first step is to add preset Gaussian white noise data to the load data set to obtain the added load data set. The preset Gaussian white noise data can be data with a normal probability distribution, an uncorrelated second-order moment and a constant first-order moment.

[0121] In the second step, based on each after-load data in the after-load data set, the following first data decomposition step is performed:

[0122] Sub-step 1, based on the added afterload data, performing the following second data decomposition step:

[0123] The first sub-step is to determine the load local maximum value set and the load local minimum value set of the added load data.

[0124] The second sub-step is to connect the load maximum value set and the load minimum value set by using a spline curve to obtain a load maximum value envelope and a load minimum value envelope. The load maximum value envelope may be an envelope formed by connecting the load maximum value sets by using a spline curve. The load minimum value envelope may be an envelope formed by connecting the load minimum value sets by using a spline curve.

[0125] The third sub-step is to determine the average of the load maximum envelope and the load minimum envelope as the load average envelope.

[0126] In the fourth sub-step, the difference between the added load data and the load average envelope is determined as the load difference data.

[0127] In a fifth sub-step, in response to determining that the load difference data satisfies a preset condition, the difference between the added load data and the load difference data is determined as the margin difference data. The preset condition may be that the sum of the number of local maximum value sets and local minimum value sets in the load difference data is equal to the number of zero crossing points or the number of differences is 1, and the mean of the maximum envelope corresponding to the local maximum value set and the minimum envelope corresponding to the local minimum value set of the load difference data is zero.

[0128] The sixth sub-step is, in response to determining that the load difference data does not meet the preset condition, determining the load difference data as the added load data, so as to perform the above-mentioned second data decomposition step again.

[0129] The seventh sub-step, in response to determining that the residual difference data is a monotonic sequence function, ends the above-mentioned first data decomposition step to obtain a residual difference data group.

[0130] Sub-step 2, in response to determining that the load margin data is not a monotonic sequence function, determining the load margin data after adding the preset Gaussian white noise as the added load data, so as to perform the above-mentioned first data decomposition step again.

[0131] The third step is to perform feature extraction processing on the obtained load difference data set to obtain a load feature vector set. The load feature vector in the above load feature vector set can represent the feature information of the load difference data set. The above feature extraction processing can be performed by extracting the nonlinear local features of the above load difference data set using a convolutional neural network model.

[0132] The fourth step is to input the load feature vector set into a bidirectional long short-term memory network to obtain a load time series feature vector set, wherein the load time series feature vector in the load time series feature vector set can be a feature vector obtained by extracting the bidirectional time series features in the load feature vector set.

[0133] The fifth step is to input the load time series feature vector set into the time series pattern attention mechanism layer to obtain a load weight feature vector set. The load weight feature vector in the load weight feature vector set may be a feature vector obtained by performing time series weighted processing on the load time series feature vector set.

[0134] The fifth step is to input the above load weight feature vector set into the time series prediction layer to obtain the load demand in the predicted time period. The above time series prediction model can be a fully connected layer.

[0135] The above-mentioned technical scheme and its related contents, combined with step "Step 107", as an inventive point of an embodiment of the present disclosure, solves the second technical problem mentioned in the background technology: "Because the load data set contains a large amount of redundant and noisy data, it is easy to cause inaccurate prediction of load demand, and the accuracy of the virtual power plant objective function and virtual constraint condition function set of the virtual power plant cluster is low, which makes the resource allocation of the virtual power plants in the virtual power plant cluster unbalanced, resulting in low system stability of the virtual power plant cluster and high damage rate of the unit equipment included." The factors that lead to low system stability of virtual power plant clusters and high damage rate of the included units are often as follows: due to the large amount of redundant and noisy data in the load data set, it is easy to cause inaccurate prediction of load demand, the virtual power plant objective function and virtual constraint condition function set of the virtual power plant cluster are of low accuracy, which makes the resource allocation of virtual power plants in the virtual power plant cluster unbalanced, and due to the problem of modal aliasing in data decomposition using the empirical mode decomposition algorithm, it is easy to cause incomplete decomposed data and the generation of noise data, and the autoregressive moving average model only predicts by focusing on the decomposed load data, and the feature information extraction of the decomposed load model is not comprehensive, resulting in inaccurate prediction of load demand. If the above factors are solved, the effect of improving the system stability of the virtual power plant cluster and reducing the damage rate of the included units can be achieved. In order to achieve this effect, the present disclosure first adds preset white noise data to the data source call information in each cycle process, which can solve the problem of modal aliasing and the problem of residual white noise in the cycle process to a certain extent, reduce the volatility and irregularity of the load data, and make the decomposed data more stable and smooth. Secondly, the obtained load difference data set is subjected to nonlinear local feature and bidirectional time series global feature extraction. The extraction of local and global features can improve the accuracy of the extracted feature vector, so as to improve the accuracy of subsequent predictions. Then, the extracted load time series feature vector set is input into the time series pattern attention mechanism layer to distinguish the importance by weighting, which can more effectively utilize the time series attributes of the load data itself and mine deeper time correlations. The time series pattern attention mechanism layer can effectively reduce the loss of historical information and give higher weight values ​​to important feature vectors, which can reduce redundant feature vectors. Finally, the load weight feature vector set is predicted, and the virtual power plant objective function and virtual power plant constraint condition set are determined by the predicted load demand to replace resources for the virtual power plant cluster, which can improve the accuracy of load demand prediction, thereby improving the accuracy of resource replacement of the virtual power plant cluster and improving the stability of the virtual power plant cluster.

[0136] Step 106: Generate resource replacement information for the virtual power plant cluster according to the virtual power plant objective function and the virtual power plant constraint function set.

[0137] In some embodiments, the execution subject may generate resource replacement information for the virtual power plant cluster according to the virtual power plant objective function and the virtual power plant constraint function set, wherein the resource replacement information may be information for replacing distributed energy resources between and within virtual power plant clusters.

[0138] As an example, the execution subject may generate resource replacement information for the virtual power plant cluster using a heuristic algorithm according to the virtual power plant objective function and the virtual power plant constraint function set, wherein the heuristic algorithm may be a genetic algorithm or a simulated annealing algorithm.

[0139] In some optional implementations of some embodiments, the generating of resource replacement information for the virtual power plant cluster according to the virtual power plant objective function and the virtual power plant constraint function set may include the following steps:

[0140] In the first step, a chaos mapping algorithm is used to determine the initialization particle group according to the virtual power plant objective function and the virtual power plant constraint function set. Each of the initialization particles in the initialization particle group may include: an initialization speed and an initialization position. The chaos mapping algorithm may be a Circle chaos mapping algorithm. Each particle in the initialization particle group may represent a feasible resource replacement scheme for the virtual power plant objective function and the virtual power plant constraint function set. The initialization speed may refer to the direction of movement of the particle from the current position to the next position, that is, the search direction of the particle group in the solution space. The initial position may be a numerical value corresponding to each parameter in the virtual power plant objective function, that is, a feasible resource replacement information.

[0141] In the second step, based on the initialized particle swarm, the following particle swarm determination steps are performed:

[0142] Sub-step 1, determining whether the initialized particle swarm satisfies the above virtual power plant constraint function set.

[0143] Sub-step 2, in response to determining that the initialized particle group satisfies the above-mentioned virtual power plant constraint function set, the initialized particle group is input into the above-mentioned virtual power plant objective function to obtain an initial fitness value set. The initial fitness value in the above-mentioned initial fitness value set can represent the degree of influence of the resource replacement information represented by the particle on the revenue of the virtual power plant. The larger the above-mentioned initial fitness value, the better the fitness of the particle represented.

[0144] Sub-step 3, according to the initial fitness value set, determine the initial target position of each initialized particle in the initialized particle group and the cluster target position of the initialized particle group, and obtain the initial target position set and the cluster target position. Among them, the initial target position in the above-mentioned initial target position set can be the position corresponding to the maximum fitness obtained by inputting the position information obtained by the particle from the initial cycle iteration to the current iteration number into the virtual power plant objective function. The above-mentioned cluster target position can be the position corresponding to the maximum fitness value in the particle group in each cycle iteration.

[0145] As an example, the execution subject may first determine the particle position group of each particle in the particle group from the start of the loop iteration to the current loop iteration number, and obtain a particle position group set. Secondly, for each particle position group in the particle position group set, determine the maximum value of the fitness value corresponding to the particle position group as the initial target position. Then, determine the fitness value set of the particle group at the current iteration number as the current particle fitness value set. Finally, determine the current particle fitness value with the largest value in the current particle fitness value set as the cluster target position.

[0146] Sub-step 4, updating the initialized particle swarm according to the initial target position set and the cluster target position to obtain an updated particle swarm.

[0147] As an example, the execution subject can use the speed update formula to update the speed of each particle in the above-mentioned initialized particle group to obtain an updated speed set. Then, use the position update formula to update the position of each particle in the above-mentioned initialized particle group to obtain an updated position set. Finally, the updated speed set and the updated position set are determined as the updated particle group. Among them, the above-mentioned speed update formula can be expressed as:

[0148]

[0149] in, represents the updated particle velocity after k cycles. c1 represents the acceleration constant, which is 2. Indicates the initial target position. represents the position of the particle of the i-th j-dimensional vector. c2 represents the acceleration constant, which is 2. Indicates the cluster target location. rand1 indicates a random number in the range [0,1]. rand2 indicates a random number in the range [0,1].

[0150] The above position update formula can be expressed as:

[0151]

[0152] Sub-step 5, inputting the updated particle swarm into the above virtual power plant objective function to obtain an updated fitness value set.

[0153] Sub-step 6, according to the initial fitness value set and the updated fitness value set, the updated particle group is screened to obtain the screened cluster position and the screened position set. The screened cluster position can be the position corresponding to the fitness value with the largest fitness value in the initial fitness value set and the updated fitness value set. The screened position in the screened position set can be the position corresponding to the fitness value with the larger value in the initial fitness value and the updated fitness value.

[0154] As an example, the above-mentioned execution entity may first compare each initial fitness value in the above-mentioned initial fitness value set with the corresponding updated fitness value in the above-mentioned updated fitness value set to obtain a comparison result set. Secondly, the comparison result set whose initial fitness value is greater than or equal to the updated fitness value is filtered out from the above-mentioned comparison result set as the target comparison result set. Thirdly, the initial fitness value set corresponding to the above-mentioned target comparison result set and the updated fitness value set corresponding to the remaining comparison result set are determined as the target fitness value set. Then, the position and speed corresponding to the above-mentioned target fitness value set are determined as the filtered position set and the filtered speed set. Finally, the position corresponding to the target fitness value with the largest value is filtered out from the above-mentioned target fitness value set as the updated cluster position.

[0155] Sub-step 7, determining the number of times the above particle swarm determination step has been executed.

[0156] Sub-step 8, in response to determining that the number of executions exceeds the preset execution threshold, generating resource replacement information according to the updated particle swarm. The resource replacement information may be the resource replacement information corresponding to the particle with the largest value in the fitness value set obtained by inputting the updated particle swarm into the virtual power plant objective function. The preset execution threshold may be a preset maximum number of executions. For example, the preset execution threshold may be 50.

[0157] As an example, the execution subject may first input the updated particle group into the virtual power plant objective function to obtain a target fitness value set. Then, the resource scheduling information within and between each virtual power plant included in the virtual power plant cluster determined by the particle corresponding to the target fitness value with the largest value in the target fitness value set is determined as the resource replacement information.

[0158] In the third step, in response to determining that the number of executions does not exceed the above-mentioned preset execution threshold, the screened particle group is determined as the initialization particle group, and the sum of the number of executions and the third preset threshold is determined as the number of executions, so as to perform the above-mentioned particle group determination step again.

[0159] The above technical scheme and its related contents, combined with step "step 107", as an inventive point of an embodiment of the present disclosure, solves the technical problem three mentioned in the background technology "Due to the inaccurate solution of the virtual power plant objective function and the virtual power plant constraint condition function set, the resource allocation in the virtual power plant cluster is unbalanced, resulting in waste of power resources generated by the virtual power plant cluster and low system stability of the virtual power plant cluster". The factors that lead to the waste of power resources generated by the virtual power plant cluster and the low system stability of the virtual power plant cluster are often as follows: due to the inaccurate solution of the virtual power plant objective function and the virtual power plant constraint condition function set, the resource allocation in the virtual power plant cluster is unbalanced, resulting in waste of power resources generated by the virtual power plant cluster and low system stability of the virtual power plant cluster, and the particle swarm optimization algorithm is used to solve the virtual power plant objective function and the virtual power plant constraint condition function set to obtain resource replacement information. However, the above conventional solutions still have the following problems: due to the random initialization in the particle swarm optimization algorithm, it is easy to cause uneven distribution of the initial position of the particle swarm, increase the waste of resources and prolong the computing resources, and the particle swarm optimization algorithm is easy to fall into the local optimal solution, so that the accuracy of the generated resource replacement information is low, and the resource scheduling of the virtual power plant cluster is not accurate. If the above factors are solved, the effect of reducing the waste of power resources generated by the virtual power plant cluster and improving the system stability of the virtual power plant cluster can be achieved. In order to achieve this effect, the present disclosure first uses the chaos mapping algorithm to initialize the particle swarm to obtain the initialized particle swarm, which can obtain uniformly distributed particle individuals and reduce the randomness and quality of the particle swarm. Secondly, the initial particle swarm is updated through the speed update formula and the position update formula. Since the position update is only related to the initial target position set and the cluster target position, the position update determines the optimal position of the particle as the position after the current iterative update, which can improve the particle swarm's optimization ability and convergence speed and to a certain extent avoid falling into the local optimal solution, reducing the waste of computing resources. Finally, by generating resource replacement information through the updated particle swarm and performing resource replacement processing on the virtual power plant cluster through the resource replacement information, the accuracy of the resource replacement information can be improved, the scheduling of differential energy resources within and between the virtual power plants included in the virtual power plant cluster can be improved, and the stability of the virtual power plant cluster can be improved.

[0160] Step 107: Perform resource replacement processing on the virtual power plant cluster according to the above resource replacement information.

[0161] In some embodiments, the execution subject may perform resource replacement processing on the virtual power plant cluster according to the resource replacement information, wherein the resource replacement may be performed within and between virtual power plants included in the virtual power plant cluster.

[0162] As an example, the above-mentioned execution entity can respond to determining that the above-mentioned resource replacement information indicates that the electricity demand of the first virtual power plant is greater than the electricity supply, and the electricity demand of the second virtual power plant is greater than the electricity supply, and send the electricity consumption scheduling of the electricity supply exceeding the electricity demand in the above-mentioned second virtual power plant to the first virtual power plant for the first virtual power plant to schedule internal resources.

[0163] The above-mentioned embodiments of the present disclosure have the following beneficial effects: the virtual power plant cluster resource replacement method of some embodiments of the present disclosure can improve the accuracy of the generated resource replacement information, reduce the consumption of power resources and improve the system stability of the virtual power plant. Specifically, the reason why the relevant power lines with transmission capacity cannot carry too much electric energy, the power line loss is high, and the stability of the power grid is low is that: the resource replacement optimization of a single virtual power plant is carried out, and the resource replacement optimization information set is integrated, only the resource replacement within a single virtual power plant is considered, and the resource replacement between virtual power plants is not considered. The accuracy of the generated resource replacement information is low, resulting in unreasonable resource replacement of the virtual power plant cluster, so that some virtual power plant resources have a large amount of idle power resources, and some virtual power plants have insufficient resources, resulting in the power lines with transmission capacity cannot carry too much electric energy, the power line loss is high, and the stability of the virtual power plant is low. Based on this, the virtual power plant cluster resource replacement method of some embodiments of the present disclosure can first obtain the distributed energy data set and load data set of the virtual power plant cluster, wherein the above-mentioned distributed energy data set includes at least one of the following: distributed power supply data, storage system, controllable load, electric vehicle charging data, and power equipment operation status information. Here, the distributed energy data set and the load data set are convenient for subsequent clustering processing. Secondly, feature data extraction is performed on the above-mentioned distributed energy data set and the above-mentioned load data set to obtain a distributed energy feature vector set and a load feature vector set. Here, representative data can be extracted from the distributed energy data set and the load data set to reduce redundant data and data volume. Again, the above-mentioned distributed energy feature vector set and the above-mentioned load feature vector set are clustered to obtain a virtual power plant data class cluster set. Here, the differences and correlations between the virtual power plants included in the virtual power plant cluster can be determined, which is convenient for the subsequent resource replacement processing of the virtual power plant cluster. Next, a resource replacement model for each virtual power plant in the above-mentioned virtual power plant cluster is generated to obtain a resource replacement model set. Here, the resource replacement performance of the virtual power plant can be more accurately grasped, which is convenient for the subsequent resource replacement processing of the virtual power plant cluster. Subsequently, according to the above-mentioned virtual power plant data class cluster set and the above-mentioned resource replacement model set, the virtual power plant objective function and virtual power plant constraint function set of the above-mentioned virtual power plant cluster are generated. Here, more accurate virtual power plant objective functions and virtual power plant constraint function sets between and within virtual power plants can be obtained, so as to generate resource replacement information later. Then, based on the above virtual power plant objective function and the above virtual power plant constraint function set, resource replacement information for the above virtual power plant cluster is generated. Here, solving the virtual power plant objective function and the virtual power plant constraint function set can obtain resource replacement information with higher accuracy, reducing the waste of power resources within and between virtual power plants. Finally, based on the above resource replacement information, the above virtual power plant cluster is subjected to resource replacement processing.Here, the rationality of resource replacement of the virtual power plant cluster can be improved, the loss of power lines and the system stability of the virtual power plant cluster can be reduced. It can be concluded that the virtual power plant cluster resource replacement method can improve the accuracy of the generated resource replacement information, reduce the consumption of power resources and improve the system stability of the virtual power plant by considering the impact of resource replacement within and between virtual power plants.

[0164] Further references Figure 2 As an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of a virtual power plant cluster resource replacement device. These device embodiments are similar to Figure 1 Corresponding to the method embodiments shown, the virtual power plant cluster resource replacement device can be specifically applied to various electronic devices.

[0165] like Figure 2 As shown, a virtual power plant cluster resource replacement device 200 includes: an acquisition unit 201, a feature data extraction unit 202, a clustering unit 203, a first generation unit 204, a second generation unit 205, a third generation unit 206 and a resource replacement unit 207. The acquisition unit 201 is configured to: acquire a distributed energy data set and a load data set of a virtual power plant cluster, wherein the distributed energy data set includes at least one of the following: distributed power source data, storage system, controllable load, electric vehicle charging data, and power equipment operation status information. The feature data extraction unit 202 is configured to: extract feature data from the distributed energy data set and the load data set to obtain a distributed energy feature vector set and a load feature vector set. The clustering unit 203 is configured to: perform clustering processing on the distributed energy feature vector set and the load feature vector set to obtain a virtual power plant data cluster set. The first generation unit 204 is configured to: generate a resource replacement model for each virtual power plant in the virtual power plant cluster to obtain a resource replacement model set. The second generation unit 205 is configured to generate a virtual power plant objective function and a virtual power plant constraint function set of the virtual power plant cluster according to the virtual power plant data cluster set and the resource substitution model set. The third generation unit 206 is configured to generate resource substitution information for the virtual power plant cluster according to the virtual power plant objective function and the virtual power plant constraint function set. The resource substitution unit 207 is configured to perform resource substitution processing on the virtual power plant cluster according to the resource substitution information.

[0166] It can be understood that the units recorded in the virtual power plant cluster resource replacement device 200 are similar to those in the reference Figure 1 Therefore, the operations, features and beneficial effects described above for the method are also applicable to the virtual power plant cluster resource replacement device 200 and the units included therein, and will not be described in detail here.

[0167] Reference below Figure 3 , which shows a structural schematic diagram of an electronic device (eg, an electronic device) 300 suitable for implementing some embodiments of the present disclosure. Figure 3 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0168] like Figure 3 As shown, the electronic device 300 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the electronic device 300 are also stored. The processing device 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0169] Typically, the following devices may be connected to the I / O interface 305: input devices 306 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 308 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 309. The communication devices 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 3 The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 3 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0170] In particular, according to some embodiments of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, some embodiments of the present disclosure include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In some such embodiments, the computer program can be downloaded and installed from the network through the communication device 309, or installed from the storage device 308, or installed from the ROM 302. When the computer program is executed by the processing device 301, the above-mentioned functions defined in the method of some embodiments of the present disclosure are executed.

[0171] It should be noted that the computer-readable medium in some embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In some embodiments of the present disclosure, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In some embodiments of the present disclosure, the computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0172] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (Hyper Text Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0173] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device. The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: obtains a distributed energy data set and a load data set of a virtual power plant cluster, wherein the distributed energy data set includes at least one of the following: distributed power source data, a storage system, a controllable load, electric vehicle charging data, and power equipment operation status information; performs feature data extraction on the distributed energy data set and the load data set to obtain a distributed energy feature vector set and a load feature vector set; performs clustering processing on the distributed energy feature vector set and the load feature vector set to obtain a virtual power plant data class cluster set; generates a resource replacement model for each virtual power plant in the virtual power plant cluster to obtain a resource replacement model set; generates a virtual power plant objective function and a virtual power plant constraint function set of the virtual power plant cluster based on the virtual power plant data class cluster set and the resource replacement model set; generates resource replacement information for the virtual power plant cluster based on the virtual power plant objective function and the virtual power plant constraint function set; performs resource replacement processing on the virtual power plant cluster based on the resource replacement information.

[0174] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0175] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0176] The units described in some embodiments of the present disclosure may be implemented by software or by hardware. The described units may also be provided in a processor, for example, may be described as: a processor including an acquisition unit, a feature data extraction unit, a clustering unit, a first generation unit, a second generation unit, a third generation unit and a resource replacement unit. Among them, the names of these units do not constitute a limitation on the unit itself in certain cases, for example, the acquisition unit may also be described as a "unit for acquiring distributed energy data sets and load data sets of a virtual power plant cluster".

[0177] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0178] The above descriptions are only some preferred embodiments of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, the above-mentioned features are replaced with the technical features with similar functions disclosed in the embodiments of the present disclosure (but not limited to) and the technical solutions formed.

Claims

1. A virtual power plant cluster resource replacement method, comprising: Obtaining a distributed energy data set and a load data set of a virtual power plant cluster, wherein the distributed energy data set includes at least one of the following: distributed power source data, storage system, controllable load, electric vehicle charging data, and power equipment operating status information; Extracting characteristic data from the distributed energy data set and the load data set to obtain a distributed energy characteristic vector set and a load characteristic vector set; Clustering the distributed energy feature vector set and the load feature vector set to obtain a virtual power plant data cluster set; Generate a resource replacement model for each virtual power plant in the virtual power plant cluster to obtain a resource replacement model set, wherein generating a resource replacement model for each virtual power plant in the virtual power plant cluster to obtain a resource replacement model set includes: For each virtual power plant in the virtual power plant cluster, the following weighted summation step is performed: In response to receiving the resource scheduling information, determining the temperature control load power of the virtual power plant; Determine a first temperature-controlled load power and a second temperature-controlled load power of the temperature-controlled load power according to a preset environmental threshold; Determining a resource replacement response direction and a resource replacement response depth of the virtual power plant according to the temperature-controlled load power, the first temperature-controlled load power, and the second temperature-controlled load power; Determining a resource replacement response time, a resource replacement response duration, and a resource replacement response rate of the virtual power plant; Determine the resource replacement response direction, the resource replacement response depth, the resource replacement response time, the resource replacement response duration, and the resource replacement response rate as a resource replacement model of the virtual power plant; Generate a virtual power plant objective function and a virtual power plant constraint function set of the virtual power plant cluster according to the virtual power plant data class cluster set and the resource replacement model set; Generate resource replacement information for the virtual power plant cluster according to the virtual power plant objective function and the virtual power plant constraint function set; According to the resource replacement information, resource replacement processing is performed on the virtual power plant cluster.

2. The method according to claim 1, wherein: The clustering process of the distributed energy feature vector set and the load feature vector set to obtain a virtual power plant data cluster set includes: Performing data standardization processing on the distributed energy feature vector set and the load feature vector set to obtain a standardized feature vector set; Determine the target cluster value corresponding to the standardized feature vector set; Filtering out the target cluster's numerical standardized feature vectors from the standardized feature vector set as an initial cluster center feature vector set; Based on the initial cluster center feature vector set, the following cluster center determination steps are performed: Determine a characteristic distance value between each residual standardized feature vector in the residual standardized feature vector set and each initial cluster center feature vector in the initial cluster center feature vector set to obtain a characteristic distance value set, wherein the residual standardized feature vector set is a feature vector set obtained by removing the initial cluster center feature vector set from the standardized feature vector set; According to the characteristic distance value group set, an initial cluster center feature vector corresponding to each remaining standardized feature vector in the remaining standardized feature vector set is determined to obtain an initial virtual power plant data class cluster set; Determine the mean of the feature vectors of each initial virtual power plant data cluster in the initial virtual power plant data cluster set as the updated cluster center feature vector, and obtain the updated cluster center feature vector set; In response to determining that each updated cluster center feature vector in the updated cluster center feature vector set is the same as an initial cluster center feature vector corresponding to the updated cluster center feature vector in the initial cluster center feature vector set, determining the initial virtual power plant data class cluster set as the virtual power plant data class cluster set; In response to determining that each updated cluster center feature vector in the updated cluster center feature vector set is different from the initial cluster center feature vector corresponding to the updated cluster center feature vector in the initial cluster center feature vector set, the updated cluster center feature vector set is determined as the initial cluster center feature vector set to perform the cluster center determination step again.

3. The method according to claim 2, wherein: The step of determining the target cluster value corresponding to the standardized feature vector set includes: Determine the arithmetic square root of the number of eigenvectors of the standardized eigenvectors included in the standardized eigenvector set as the termination cluster value; Setting the initial cluster value of the target cluster value to a first preset threshold; Based on the initial cluster values, perform the following cluster average evaluation coefficient screening steps: Perform clustering on the standardized feature vector set to obtain feature vector clusters with an initial cluster value; For each feature vector cluster in the initial cluster number feature vector clusters, perform the following clustering evaluation coefficient generation steps: Determine the average intra-cluster distance value of the standardized feature vector set included in each feature vector cluster in the initial cluster value feature vector clusters and the feature vector corresponding to the cluster center of each feature vector cluster as the average intra-cluster distance value and the central feature vector, and obtain the average intra-cluster distance value set and the central feature vector set; Determine the inter-cluster distance value of the central feature vector corresponding to the feature vector cluster and each remaining central feature vector in the remaining central feature vector set to obtain an inter-cluster distance value set, wherein the remaining central feature vector set is a feature vector set obtained by removing the central feature vector from the central feature vector set corresponding to the initial cluster value feature vector clusters; According to the inter-cluster distance value set, the remaining central feature vector set is screened to obtain the target remaining central feature vector; Generate a clustering evaluation coefficient based on the average distance value within the cluster and the average value within the cluster corresponding to the target residual center feature vector; Determine the average value of each clustering evaluation coefficient obtained as the clustering average evaluation coefficient; Adding the cluster average evaluation coefficient to the preset cluster average evaluation coefficient list to obtain the cluster average evaluation coefficient list; The sum of the initial cluster value and the second preset threshold is determined as the updated cluster value; In response to determining that the updated cluster value is greater than or equal to the termination cluster value, a cluster average evaluation coefficient that meets a preset condition is selected from the obtained cluster average evaluation coefficient list as a target cluster average evaluation coefficient, and an initial cluster value corresponding to the target cluster average evaluation coefficient is determined as the target cluster value; In response to determining that the updated cluster value is less than the termination cluster value, the updated cluster value is determined as the initial cluster value to perform the cluster average evaluation coefficient screening step again.

4. The method according to claim 1, wherein: The step of generating a virtual power plant objective function and a virtual power plant constraint function set of the virtual power plant cluster according to the virtual power plant data class cluster set and the resource replacement model set includes: Obtaining the power acquisition value, load power and power consumption value of each virtual power plant data cluster in the virtual power plant data cluster set to obtain a power acquisition value set, a load power set and a power consumption value set; Determine the power acquisition value function of the virtual power plant cluster by adding the product of each power acquisition value in the power acquisition value set and the load power corresponding to the power acquisition value in the load power set; Determine the power consumption value function of the virtual power plant cluster by adding the product of each power consumption value in the power consumption value set and the load power corresponding to the power consumption value in the load power set; In response to determining that there is at least one resource replacement response depth less than a preset resource replacement response depth threshold in the resource replacement response depth set included in the resource replacement model set, determining the at least one resource replacement response depth as a target resource replacement response depth set; Determine a replacement range penalty function of each virtual power plant corresponding to the target resource replacement response depth set; In response to determining that there is at least one resource replacement response rate greater than a preset resource replacement response rate threshold in the resource replacement response rate set included in the resource replacement model set, determining the at least one resource replacement response rate as a target resource replacement response rate set; Determine a response rate penalty function of each virtual power plant corresponding to the target resource replacement response rate set; The power acquisition value function, the power consumption value function, the replacement range penalty function and the response rate penalty function are subtracted to obtain a virtual power plant objective function.

5. The method according to claim 4, wherein: The step of generating a virtual power plant objective function and a virtual power plant constraint function set of the virtual power plant cluster according to the virtual power plant data class cluster set and the resource replacement model set further includes: Performing load forecasting on the load data set to obtain load demand in a forecasting time period; Determining the accumulated sum of the load power sets as the target load power of the virtual power plant cluster; Generate a power supply and demand balance constraint function of the virtual power plant cluster according to the difference between the target load power and the load demand in the forecast time period and a preset power grid deviation threshold; Generate an equipment load constraint condition function of the virtual power plant cluster according to each load power in the load power set, a first preset load power threshold, and a second preset load power threshold; Determine the equipment start and stop constraint condition function of the cluster unit equipment set of the virtual power plant cluster; Determine a response rate constraint condition function of the virtual power plant cluster according to a ramp rate corresponding to each load power in the load power set and a resource replacement response rate set corresponding to the resource replacement model set; The power supply and demand balance constraint condition function, the equipment load constraint condition function, the equipment start and stop constraint condition function and the response rate constraint condition function are determined as a virtual power plant constraint function set of the virtual power plant cluster.

6. A virtual power plant cluster resource replacement device, comprising: An acquisition unit is configured to acquire a distributed energy data set and a load data set of a virtual power plant cluster, wherein the distributed energy data set includes at least one of the following: distributed power source data, storage system, controllable load, electric vehicle charging data, and power equipment operation status information; A feature data extraction unit is configured to extract feature data from the distributed energy data set and the load data set to obtain a distributed energy feature vector set and a load feature vector set; A clustering unit is configured to perform clustering processing on the distributed energy feature vector set and the load feature vector set to obtain a virtual power plant data cluster set; The first generating unit is configured to generate a resource replacement model for each virtual power plant in the virtual power plant cluster to obtain a resource replacement model set, wherein the generating of the resource replacement model for each virtual power plant in the virtual power plant cluster to obtain the resource replacement model set includes: for each virtual power plant in the virtual power plant cluster, performing the following weighted summation step: in response to receiving resource scheduling information, determining the temperature control load power of the virtual power plant; determining a first temperature control load power and a second temperature control load power of the temperature control load power according to a preset environmental threshold; determining a resource replacement response direction and a resource replacement response depth of the virtual power plant according to the temperature control load power, the first temperature control load power and the second temperature control load power; determining a resource replacement response time, a resource replacement response duration and a resource replacement response rate of the virtual power plant; determining the resource replacement response direction, the resource replacement response depth, the resource replacement response time, the resource replacement response duration and the resource replacement response rate as the resource replacement model of the virtual power plant; A second generating unit is configured to generate a virtual power plant objective function and a virtual power plant constraint function set of the virtual power plant cluster according to the virtual power plant data class cluster set and the resource substitution model set; A third generating unit is configured to generate resource replacement information for the virtual power plant cluster according to the virtual power plant objective function and the virtual power plant constraint function set; The resource replacement unit is configured to perform resource replacement processing on the virtual power plant cluster according to the resource replacement information.

7. An electronic device comprising: one or more processors; a storage device having one or more programs stored thereon, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 5.

8. A computer readable medium having a computer program stored thereon, wherein: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images