Method and device for determining power generation strategy and computer program product

Abstract

The invention discloses a power generation strategy determination method and device and a computer program product, and relates to the technical field of auxiliary decision making, and the power generation strategy determination method comprises the steps: obtaining different types of power generation strategy data; obtaining a power demand prediction result output by a first prediction model based on the different types of power generation strategy data, and obtaining a power generation output prediction result output by a second prediction model based on the different types of power generation strategy data, the first prediction model and the second prediction model are obtained by training based on historical power generation strategy data; and determining a target power generation strategy according to the power demand prediction result and the power generation output prediction result. By adopting the technical scheme, the problem of how to accurately determine the power generation strategy is solved.

Description

Technical Field

[0001] The present application relates to the technical field of auxiliary decision making, and in particular, to a method, device and computer program product for determining a power generation strategy. Background Art

[0002] With the continuous development and improvement of the power market, power generation companies need to more accurately grasp market demand and formulate scientific power production plans to improve economic benefits and market competitiveness. Therefore, an auxiliary decision-making platform based on power market demand came into being.

[0003] At present, there are some decision-making support systems on the market, but most of them focus on production scheduling and cost control within power generation enterprises, lacking in-depth analysis and prediction of power market demand. At the same time, these systems often use traditional mathematical models and algorithms, which are difficult to adapt to the complex and changeable characteristics of the power market. Therefore, it is impossible to accurately predict the power market demand, making it difficult for power generation enterprises to determine accurate power generation strategies.

[0004] Therefore, in the related art, there is a problem of how to accurately determine the power generation strategy.

[0005] With regard to the problem of how to accurately determine the power generation strategy in the related technology, no effective solution has been proposed yet.

[0006] Therefore, it is necessary to improve the related technology to overcome the above-mentioned defects in the related technology. Summary of the invention

[0007] The embodiments of the present application provide a method, device and computer program product for determining a power generation strategy, so as to at least solve the problem of how to accurately determine a power generation strategy in the related art.

[0008] According to one aspect of an embodiment of the present application, a method for determining a power generation strategy is provided, comprising: obtaining different types of power generation strategy data; obtaining a power demand forecast result output by a first forecasting model based on the different types of power generation strategy data, and obtaining a power generation output forecast result output by a second forecasting model based on the different types of power generation strategy data, wherein both the first forecasting model and the second forecasting model are trained based on historical power generation strategy data; and determining a target power generation strategy based on the power demand forecast result and the power generation output forecast result.

[0009] In an exemplary embodiment, after acquiring different types of power generation strategy data, the method further includes: performing noise reduction processing on the different types of power generation strategy data in the following manner: for each type of power generation strategy data, setting multiple centroids, respectively calculating the distance between each data and the multiple centroids, and assigning each data to a cluster where the nearest centroid is located; updating the centroid according to the average value of multiple data corresponding to each cluster; and when it is determined that the updated value of the centroid is less than a first preset value, merging clusters whose data volume is less than a second preset value.

[0010] In an exemplary embodiment, when it is determined that the updated value of the centroid is less than the first preset value, after merging clusters whose data volume is less than the second preset value, the method further includes: performing data cleaning on the different types of power generation strategy data in the following manner: calculating the average distance of each cluster, wherein the average distance represents the average value of the distances between multiple data corresponding to a cluster and the centroid; determining data whose distance from the centroid is greater than a third preset value as abnormal data; and cleaning the abnormal data.

[0011] In an exemplary embodiment, before obtaining the power demand forecast results output by the first forecasting model based on the different types of power generation strategy data, and obtaining the power generation output forecast results output by the second forecasting model based on the different types of power generation strategy data, the method further includes: normalizing each type of power generation strategy data by the following formula:

[0012]

[0013] Among them, X' represents the normalized data, X represents the initial data, and X min is the minimum value of each type of power generation strategy data, X max Representing the maximum value of each type of power generation strategy data; converting all normalized data in each type of power generation strategy data into a data set, and converting the data set into a vector set; and inputting the vector set into the first prediction model and the second prediction model respectively.

[0014] In an exemplary embodiment, the different types of power generation strategy data include historical power demand data, and the power demand is predicted by a first prediction model to obtain a power demand prediction result, including: the first prediction model predicts short-term power demand by the following formula:

[0015]

[0016] Among them, S t represents the short-term electricity demand forecast value, X x represents the historical electricity demand data, Xx(t-i) Represents the historical electricity demand data X at time point ti x The value in n represents the moving window size of the time series, and w i represents the weight of the observation value; the first prediction model predicts the long-term electricity demand through the following formula:

[0017]

[0018] Among them, Y t It is represented by the long-term electricity demand forecast value at the current time t, c is a constant term, which represents the baseline level of the model, is the autoregressive part, which represents the relationship between the current value and the value at p time points in the past. p is the number of autoregressive terms, and φ i is the autoregressive coefficient, which indicates the influence of the lag period on the current value. is the moving average part, which represents the relationship between the current value and the random error of the past q time points, q is the number of moving average items, θ j is the moving average coefficient, which indicates the influence of the random error in the lag period on the current value, ε t is a random error term, indicating random fluctuation; the power demand forecast result is determined according to the short-term power demand forecast value and the long-term power demand forecast value.

[0019] In an exemplary embodiment, the multiple types of data include meteorological data, operating data of power generation equipment, and historical power data, and the power generation output is predicted by the second prediction model to obtain the power generation output prediction result, including: the second prediction model obtains the power generation output prediction result by the following formula:

[0020]

[0021] Among them, y i represents the power generation output prediction result output by the output layer, w hj represents the weight from the hidden layer to the output layer, σ represents the activation function, m represents the number of hidden layers, a h represents the input of the hidden layer neuron, w th represents the weight from the input layer to the hidden layer, x i represents the output of the input layer, b h represents the bias of the hidden layer, n represents the number of input layers, and the data input to the input layer at least includes the meteorological data, the operating status of the power generation equipment and the historical power data information.

[0022] In an exemplary embodiment, the method further includes: calculating the output layer error by the following formula: Among them, e j represents the output layer error, represents the expected output of the output layer, y j Represents the actual output of the output layer; the hidden layer error is calculated by the following formula: Among them, δ h Represents the error of the hidden layer, l represents the number of output layers, and σ' represents the derivative of the activation function; the weights from the input layer to the hidden layer are updated by the following formula: Among them, η represents the learning rate, represents the updated weight, Represents the weight before updating.

[0023] According to another aspect of an embodiment of the present application, a device for determining a power generation strategy is also provided, including: an acquisition module for acquiring different types of power generation strategy data; a prediction module for acquiring a power demand prediction result output by a first prediction model based on the different types of power generation strategy data, and acquiring a power generation output prediction result output by a second prediction model based on the different types of power generation strategy data, wherein both the first prediction model and the second prediction model are trained based on historical power generation strategy data; and a determination module for determining a target power generation strategy based on the power demand prediction result and the power generation output prediction result.

[0024] According to another aspect of the embodiments of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned method for determining the power generation strategy when running.

[0025] According to another aspect of an embodiment of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the method for determining the power generation strategy through the computer program.

[0026] According to another aspect of the embodiments of the present application, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps of the method described in each embodiment of the present application are implemented.

[0027] Through this application, different types of power generation strategy data can be obtained, and after processing, different types of power generation strategy data can be input into different prediction models to obtain power demand prediction results and power generation output prediction results respectively, and then the target power generation strategy can be determined according to the power demand prediction results and power generation output prediction results. This solves the problem of how to accurately determine the power generation strategy in the related art and achieves the effect of improving the accuracy of the power generation strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0029] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0030] Figure 1 It is a hardware structure block diagram of a computer terminal of a method for determining a power generation strategy in an embodiment of the present application;

[0031] Figure 2 is a flow chart of a method for determining a power generation strategy according to an embodiment of the present application;

[0032] Figure 3 is a schematic diagram of a decision support platform according to an embodiment of the present application;

[0033] Figure 4 It is a structural block diagram of a device for determining a power generation strategy according to an embodiment of the present application. DETAILED DESCRIPTION

[0034] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0035] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0036] The method embodiments provided in the embodiments of the present application can be executed in a computer terminal or a similar computing device. Taking running on a computer terminal as an example, Figure 1 1 is a hardware structure block diagram of a computer terminal for determining a power generation strategy according to an embodiment of the present application. Figure 1 As shown, the computer terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor (Central Processing Unit, MCU) or a programmable logic device (Field Programmable Gate Array, FPGA)) and a memory 104 for storing data, wherein the above-mentioned computer terminal may also include a transmission device 106 and an input and output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above-mentioned computer terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0037] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the method for determining the power generation strategy in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories can be connected to the computer terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0038] A wireless network provided by a communication provider of a computer terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (Radio Frequency, referred to as RF) module, which is used to communicate with the Internet wirelessly.

[0039] In this embodiment, a method for determining a power generation strategy is provided. Figure 2is a flow chart of a method for determining a power generation strategy according to an embodiment of the present application, such as Figure 2 As shown, the process includes the following steps:

[0040] Step S202, obtaining different types of power generation strategy data;

[0041] Optionally, in the above step S202, different types of power generation strategy data include, for example, historical power demand data obtained from the power market, collected meteorological data, monitoring of the operating status of power generation equipment, historical power data information, etc.

[0042] Step S204, obtaining a power demand forecast result output by a first forecasting model based on the different types of power generation strategy data, and obtaining a power generation output forecast result output by a second forecasting model based on the different types of power generation strategy data, wherein both the first forecasting model and the second forecasting model are trained based on historical power generation strategy data;

[0043] Optionally, in the above step S204, the first prediction model can predict short-term and long-term changes in electricity demand by performing machine learning on a vector set of historical power generation strategy data, and the second prediction model can perform machine learning based on meteorological data and historical power generation records to predict the output of various power generation methods, thereby providing a basis for power generation decisions.

[0044] Step S206: determining a target power generation strategy according to the power demand prediction result and the power generation output prediction result.

[0045] Optionally, in the above step S206, power production scheduling can be optimized and costs can be reduced based on power demand forecasts and power generation output capacity, and priority recommendations for different power generation methods can be provided. The choice of power generation methods can be flexibly adjusted, and strategies that flexibly respond to market changes can be provided based on power market price trends to achieve decision support for rapid start and stop of power generation units.

[0046] Through the above steps, different types of power generation strategy data can be obtained, and after processing, different types of power generation strategy data are input into different prediction models to obtain power demand prediction results and power generation output prediction results respectively, and then the target power generation strategy is determined according to the power demand prediction results and the power generation output prediction results. This solves the problem of how to accurately determine the power generation strategy in the related art and achieves the effect of improving the accuracy of the power generation strategy.

[0047] In an exemplary embodiment, after acquiring different types of power generation strategy data, the method further includes: performing noise reduction processing on the different types of power generation strategy data in the following manner: for each type of power generation strategy data, setting multiple centroids, respectively calculating the distance between each data and the multiple centroids, and assigning each data to a cluster where the nearest centroid is located; updating the centroid according to the average value of multiple data corresponding to each cluster; and when it is determined that the updated value of the centroid is less than a first preset value, merging clusters whose data volume is less than a second preset value.

[0048] Optionally, in the above embodiment, the calculation process of performing noise reduction processing on the different types of power generation strategy data is as follows:

[0049] Set the center of mass c j , and then calculate each data information x i With the centroid c j Distance:

[0050]

[0051] Where n is the number of data messages, x ik Is data information x i The kth eigenvalue of jk is the center of mass c j The kth eigenvalue of ;

[0052] In each iteration, the data information x i Assign to the nearest centroid c j The cluster where it is located is calculated as:

[0053]

[0054] Represented as each data information x i will be assigned to minimize the distance d(x i ,c j )’s centroid c j Inside;

[0055] Update the centroid of each cluster:

[0056]

[0057] Among them, N j is the amount of data information in cluster j, C j is all the data information in cluster j;

[0058] Repeat the assignment and update process until the centroid no longer changes or changes very little;

[0059] After clustering is completed, clusters with smaller sample sizes are merged to reduce the interference of small clusters on the overall data information caused by noise.

[0060] In an exemplary embodiment, when it is determined that the updated value of the centroid is less than the first preset value, after merging clusters whose data volume is less than the second preset value, the method further includes: performing data cleaning on the different types of power generation strategy data in the following manner: calculating the average distance of each cluster, wherein the average distance represents the average value of the distances between multiple data corresponding to a cluster and the centroid; determining data whose distance from the centroid is greater than a third preset value as abnormal data; and cleaning the abnormal data.

[0061] Optionally, in the above embodiment, the calculation process of performing data cleaning on the different types of power generation strategy data is as follows:

[0062]

[0063] Among them, F is the outlier factor, D is the collected data information, and N j is the number of samples in the jth cluster group, d(x i ,c j ) is the data information x i The distance to the jth cluster center, and C j is the jth cluster center;

[0064] For the distance away from the cluster center, it is set to average the distance of the data information from the center position, and the data results processed by the cluster analysis algorithm are calculated:

[0065]

[0066] Among them, C j is the jth cluster center. If the distance between the data and the cluster center is greater than 3d, the data is judged to be abnormal data and then cleaned up.

[0067] In an exemplary embodiment, before obtaining the power demand forecast results output by the first forecasting model based on the different types of power generation strategy data, and obtaining the power generation output forecast results output by the second forecasting model based on the different types of power generation strategy data, the method further includes: normalizing each type of power generation strategy data by the following formula:

[0068]

[0069] Among them, X' represents the normalized data, X represents the initial data, and X min is the minimum value of each type of power generation strategy data, X maxRepresenting the maximum value of each type of power generation strategy data; converting all normalized data in each type of power generation strategy data into a data set, and converting the data set into a vector set; and inputting the vector set into the first prediction model and the second prediction model respectively.

[0070] Optionally, in the above embodiment, the process of normalizing different types of power generation strategy data and then converting them into vector sets is as follows:

[0071] Set the historical power demand data to X x , weather data is X q , the operating state of the power generation equipment is X d and historical power data informationX l ;

[0072] The collected historical power demand data is X x , weather data is X q , the operating state of the power generation equipment is X d and historical power data informationX l To convert a collection:

[0073] First, normalize the data to convert it to the same scale:

[0074]

[0075] Among them, X' represents the normalized data, X represents the initial data, and X min is the minimum value in the initial data, X max is the maximum value in the initial data;

[0076] Then convert the normalized data into a set:

[0077] X x =[X' x1 ,X' x2 ,…,X' xn ],

[0078] X q =[X' q1 ,X' q2 ,…,X' qn ],

[0079] X d =[X' d1 ,X' d2 ,…,X' dn ],

[0080] X l =[X l '1,X l'2,…,X l ' n ],

[0081] Convert a data set into a vector set:

[0082]

[0083]

[0084] In an exemplary embodiment, the different types of power generation strategy data include historical power demand data, and the power demand is predicted by a first prediction model to obtain a power demand prediction result, including: the first prediction model predicts short-term power demand by the following formula:

[0085]

[0086] Among them, S t represents the short-term electricity demand forecast value, X x represents the historical electricity demand data, X x(t-i) Represents the historical electricity demand data X at time point ti x The value in n represents the moving window size of the time series, and w i represents the weight of the observation value; the first prediction model predicts the long-term electricity demand through the following formula:

[0087]

[0088] Among them, Y t It is represented by the long-term electricity demand forecast value at the current time t, c is a constant term, which represents the baseline level of the model, is the autoregressive part, which represents the relationship between the current value and the value at the past p time points, p is the number of autoregressive terms, φ i is the autoregressive coefficient, which indicates the influence of the lag period on the current value. is the moving average part, which represents the relationship between the current value and the random error of the past q time points, q is the number of moving average items, θ j is the moving average coefficient, which indicates the influence of the random error in the lag period on the current value, ε t is a random error term, indicating random fluctuation; the power demand forecast result is determined according to the short-term power demand forecast value and the long-term power demand forecast value.

[0089] In an exemplary embodiment, the multiple types of data include meteorological data, operating data of power generation equipment, and historical power data, and the power generation output is predicted by the second prediction model to obtain the power generation output prediction result, including: the second prediction model obtains the power generation output prediction result by the following formula:

[0090]

[0091] Among them, y i represents the power generation output prediction result output by the output layer, w hj represents the weight from the hidden layer to the output layer, σ represents the activation function, m represents the number of hidden layers, a h represents the input of the hidden layer neuron, w th represents the weight from the input layer to the hidden layer, x i represents the output of the input layer, b h represents the bias of the hidden layer, n represents the number of input layers, and the data input to the input layer at least includes the meteorological data, the operating status of the power generation equipment and the historical power data information.

[0092] In an exemplary embodiment, the method further includes: calculating the output layer error by the following formula: Among them, e j represents the output layer error, represents the expected output of the output layer, y j Represents the actual output of the output layer; the hidden layer error is calculated by the following formula: Among them, δ h Represents the error of the hidden layer, l represents the number of output layers, and σ' represents the derivative of the activation function; the weights from the input layer to the hidden layer are updated by the following formula: Among them, η represents the learning rate, represents the updated weight, Represents the weight before updating.

[0093] Obviously, the above-described embodiments are only a part of the embodiments of the present application, not all of the embodiments. In order to better understand the above-mentioned method for determining the power generation area, the above-mentioned process is described below in conjunction with the embodiments, but it is not intended to limit the technical solutions of the embodiments of the present application, specifically:

[0094] In an optional embodiment, the method for determining the power generation strategy of the present application can be applied to Figure 3 The electric energy production auxiliary decision-making platform shown in the figure specifically includes:

[0095] The Internet communication module is used to establish a communication connection between the auxiliary decision-making platform and various places, to query the data information of various places, and to transmit the collected data.

[0096] The data acquisition module is used to collect and query data information from various places. The data information collected and queried by the data acquisition module includes obtaining historical power demand data from the power market; collecting meteorological data for predicting the power generation capacity of renewable energy; monitoring the operating status of power generation equipment, including fault alarms and maintenance requirements, and obtaining historical power data information from various places, and the collected data information is transmitted through the Internet communication module.

[0097] The data analysis module is used to process the data information collected by the data acquisition module, including preprocessing and data conversion of the data information. The preprocessing includes denoising, cleaning, format conversion and unification of the data information. The data conversion is used to convert the preprocessed data information into the form of a data set, and establish a vector set based on the data set.

[0098] The data prediction module is used to predict short-term and long-term changes in electricity demand by performing machine learning model analysis on vector sets, and to predict the output of various power generation methods based on weather data and historical power generation records, and to combine the analysis of power market price trends to provide a basis for decision-making.

[0099] The decision support module is used to optimize power production scheduling and reduce costs based on demand forecasts and power generation capacity, and provide priority recommendations for different power generation methods. It can flexibly adjust the choice of power generation methods and provide strategies that flexibly respond to market changes based on power market price trends, thereby achieving decision support for rapid start and stop of units and long-distance transmission and purchase of electricity.

[0100] Optionally, the above-mentioned power production auxiliary decision platform can communicate data with other systems such as the power dispatch center and the market trading platform or subsystems in various regions. The auxiliary decision platform can also use the database to store historical data and real-time data and perform big data analysis. Further, the auxiliary decision platform can also use cloud computing platforms and edge cloud computing platforms to provide flexible computing resources to complete complex analysis tasks.

[0101] Optionally, the process of the decision support module of the above-mentioned electric energy production auxiliary decision platform making a decision and outputting the electric energy production includes, for example:

[0102] The historical electricity demand data is X x , weather data is X q , the operating state of the power generation equipment is X d and historical power data informationX l The data information is processed by information gain, and the collected data information is converted into a data set, and:

[0103]

[0104] Among them, H(N) is the entropy of data set N, H(N|A) is the entropy of data set N under the condition of optimal feature mean A, H(A) is the entropy of information gain A, λ is the adjustment coefficient, and complexity(A) is the complexity of the feature;

[0105] The entropy of the data set N is used to calculate the uncertainty of the data set, and the calculation formula is as follows:

[0106]

[0107] Among them, z represents the number of categories, p i is the probability of the i-th category; z is used to divide the data set N, and split the data set according to the information gain to form subsets;

[0108] Compare the information gains of all features, and select the feature with the largest information gain as the split feature of the current node; divide the data set into subsets according to the selected features, and each subset corresponds to a value of the feature; repeat the above process for each subset until the stopping condition is met; and the stopping condition is that the number of samples in the node is less than 10 data information. The above process can determine the nodes of the decision tree, and then realize the construction of the decision tree.

[0109] After the decision tree is constructed, it can be pruned by using complexity penalty. The specific formula is as follows:

[0110] C total =C crror +α×C sizc +β×C depth ,

[0111] Among them, C crror is the error rate of the decision tree; C sizc is the number of leaf nodes of the decision tree; C depth is the maximum depth of the decision tree; α is the penalty coefficient for controlling the tree size; β is the penalty coefficient for controlling the tree depth.

[0112] Through the above-mentioned embodiments, it is possible to integrate electricity market data, internal data of power generation enterprises, and information from external data sources, use advanced algorithms and models to accurately predict market demand, and provide comprehensive decision support for power generation enterprises. Using advanced machine learning algorithms and deep learning technologies, a market demand forecasting model is constructed to achieve accurate forecasts of electricity market demand, including forecasts of short-term and long-term electricity demand changes, the output of various power generation methods, and electricity market price trends. This can then optimize power production scheduling, flexibly adjust power generation methods, and select strategies that can flexibly respond to market changes. By implementing the present invention, power generation enterprises can achieve accurate forecasts of electricity market demand, improve economic benefits and market competitiveness, and provide strong support for the digital transformation and intelligent upgrading of enterprises.

[0113] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of each embodiment of the present application.

[0114] In this embodiment, a device for determining a power generation strategy is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0115] Figure 4 : is a structural block diagram of a device for determining a power generation strategy according to an embodiment of the present application, the device comprising:

[0116] An acquisition module 42, used to acquire different types of power generation strategy data;

[0117] A prediction module 44 is used to obtain a power demand prediction result output by a first prediction model based on the different types of power generation strategy data, and to obtain a power generation output prediction result output by a second prediction model based on the different types of power generation strategy data, wherein both the first prediction model and the second prediction model are trained based on historical power generation strategy data;

[0118] The determination module 46 is used to determine a target power generation strategy according to the power demand prediction result and the power generation output prediction result.

[0119] Through the above device, different types of power generation strategy data can be obtained, and after processing, different types of power generation strategy data can be input into different prediction models to obtain power demand prediction results and power generation output prediction results respectively, and then the target power generation strategy can be determined according to the power demand prediction results and power generation output prediction results. This solves the problem of how to accurately determine the power generation strategy in the related technology and achieves the effect of improving the accuracy of the power generation strategy.

[0120] In an exemplary embodiment, the acquisition module 42 is also used to: perform noise reduction processing on the different types of power generation strategy data in the following manner: for each type of power generation strategy data, set multiple centroids, calculate the distance between each data and the multiple centroids respectively, and assign each data to the cluster where the nearest centroid is located; update the centroid according to the average value of multiple data corresponding to each cluster; when it is determined that the updated value of the centroid is less than a first preset value, merge the clusters whose data amount is less than a second preset value.

[0121] In an exemplary embodiment, the acquisition module 42 is also used to: perform data cleaning on the different types of power generation strategy data in the following manner: calculate the average distance of each cluster, wherein the average distance represents the average value of the distances between multiple data corresponding to a cluster and the centroid; determine data whose distance from the centroid is greater than a third preset value as abnormal data; and clean the abnormal data.

[0122] In an exemplary embodiment, the prediction module 44 is further used to normalize each type of power generation strategy data using the following formula:

[0123]

[0124] Among them, X' represents the normalized data, X represents the initial data, and X min is the minimum value of each type of power generation strategy data, X max Representing the maximum value of each type of power generation strategy data; converting all normalized data in each type of power generation strategy data into a data set, and converting the data set into a vector set; and inputting the vector set into the first prediction model and the second prediction model respectively.

[0125] In an exemplary embodiment, the different types of power generation strategy data include historical power demand data, and the prediction module 44 is further used to predict short-term power demand by the following formula:

[0126]

[0127] Among them, S t represents the short-term electricity demand forecast value, X x represents the historical electricity demand data, X x(t-i) Represents the historical electricity demand data X at time point ti x The value in n represents the moving window size of the time series, and w i represents the weight of the observation value; the first prediction model predicts the long-term electricity demand through the following formula:

[0128]

[0129] Among them, Y t It is represented by the long-term electricity demand forecast value at the current time t, c is a constant term, which represents the baseline level of the model, is the autoregressive part, which represents the relationship between the current value and the value at p time points in the past. p is the number of autoregressive terms, and φ i is the autoregressive coefficient, which indicates the influence of the lag period on the current value. is the moving average part, which represents the relationship between the current value and the random error of the past q time points, q is the number of moving average items, θ j is the moving average coefficient, which indicates the influence of the random error in the lag period on the current value, ε t is a random error term, indicating random fluctuation; the power demand forecast result is determined according to the short-term power demand forecast value and the long-term power demand forecast value.

[0130] In an exemplary embodiment, the multiple types of data include meteorological data, operating data of power generation equipment, and historical power data. The prediction module 44 is further used to obtain the power generation output prediction result by the following formula:

[0131]

[0132] Among them, y i represents the power generation output prediction result output by the output layer, w hj represents the weight from the hidden layer to the output layer, σ represents the activation function, m represents the number of hidden layers, a h represents the input of the hidden layer neuron, w th represents the weight from the input layer to the hidden layer, x i represents the output of the input layer, b h represents the bias of the hidden layer, n represents the number of input layers, and the data input to the input layer at least includes the meteorological data, the operating status of the power generation equipment and the historical power data information.

[0133] In an exemplary embodiment, the prediction module 44 is further used to calculate the output layer error by the following formula: Among them, e j represents the output layer error, represents the expected output of the output layer, y j Represents the actual output of the output layer; the hidden layer error is calculated by the following formula: Among them, δ h Represents the error of the hidden layer, l represents the number of output layers, and σ' represents the derivative of the activation function; the weights from the input layer to the hidden layer are updated by the following formula: Among them, η represents the learning rate, represents the updated weight, Represents the weight before updating.

[0134] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0135] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

[0136] S1, obtain different types of power generation strategy data;

[0137] S2, obtaining a power demand forecast result output by a first forecasting model based on the different types of power generation strategy data, and obtaining a power generation output forecast result output by a second forecasting model based on the different types of power generation strategy data, wherein both the first forecasting model and the second forecasting model are trained based on historical power generation strategy data;

[0138] S3, determining a target power generation strategy according to the power demand prediction result and the power generation output prediction result.

[0139] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0140] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0141] An embodiment of the present application further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0142] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0143] S1, obtain different types of power generation strategy data;

[0144] S2, obtaining a power demand forecast result output by a first forecasting model based on the different types of power generation strategy data, and obtaining a power generation output forecast result output by a second forecasting model based on the different types of power generation strategy data, wherein both the first forecasting model and the second forecasting model are trained based on historical power generation strategy data;

[0145] S3, determining a target power generation strategy according to the power demand prediction result and the power generation output prediction result.

[0146] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0147] An embodiment of the present application further provides a computer program product, comprising a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program product, and when the computer program is executed by a processor, the steps of the method described in each embodiment of the present application are implemented.

[0148] Optionally, in this embodiment, the above computer program may be configured to implement the following steps when executed by a processor:

[0149] S1, obtain different types of power generation strategy data;

[0150] S2, obtaining a power demand forecast result output by a first forecasting model based on the different types of power generation strategy data, and obtaining a power generation output forecast result output by a second forecasting model based on the different types of power generation strategy data, wherein both the first forecasting model and the second forecasting model are trained based on historical power generation strategy data;

[0151] S3, determining a target power generation strategy according to the power demand prediction result and the power generation output prediction result.

[0152] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0153] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0154] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for determining a power generation strategy, characterized in that: include: Obtain data on different types of power generation strategies; Obtaining a power demand forecast result output by a first forecasting model based on the different types of power generation strategy data, and obtaining a power generation output forecast result output by a second forecasting model based on the different types of power generation strategy data, wherein both the first forecasting model and the second forecasting model are trained based on historical power generation strategy data; A target power generation strategy is determined according to the power demand forecast result and the power generation output forecast result.

2. The method according to claim 1, characterized in that After acquiring different types of power generation strategy data, the method further includes: The noise reduction process of the different types of power generation strategy data is performed in the following manner: For each type of power generation strategy data, multiple centroids are set, the distances between each data and the multiple centroids are calculated respectively, and each data is assigned to a cluster where the nearest centroid is located; Update the centroid according to the average value of multiple data corresponding to each cluster; When it is determined that the updated value of the centroid is less than the first preset value, clusters having data amounts less than the second preset value are merged.

3. The method according to claim 2, characterized in that When it is determined that the updated value of the centroid is less than the first preset value, after merging the clusters whose data amount is less than the second preset value, the method further includes: performing data cleaning on the different types of power generation strategy data in the following manner: Calculate the average distance of each cluster, where the average distance represents the average value of the distances between multiple data corresponding to a cluster and the centroid; Determine data whose distance from the centroid is greater than a third preset value as abnormal data; The abnormal data is cleaned up.

4. The method according to claim 1, characterized in that Before obtaining the power demand forecast results output by the first forecasting model based on the different types of power generation strategy data, and obtaining the power generation output forecast results output by the second forecasting model based on the different types of power generation strategy data, the method further includes: normalizing each type of power generation strategy data by the following formula: Among them, X' represents the normalized data, X represents the initial data, and X min is the minimum value of each type of power generation strategy data, X max Indicates the maximum value of each type of power generation strategy data; Convert all normalized data in each type of power generation strategy data into a data set, and convert the data set into a vector set; The vector set is input into the first prediction model and the second prediction model respectively.

5. The method according to claim 1, characterized in that The different types of power generation strategy data include historical power demand data, and the power demand is predicted by the first prediction model to obtain a power demand prediction result, including: The first prediction model predicts short-term electricity demand by the following formula: Among them, S t represents the short-term electricity demand forecast value, X x represents the historical electricity demand data, X x(t-i) Represents the historical electricity demand data X at time point ti x The value in n represents the moving window size of the time series, and w i represents the weight of the observation; The first prediction model predicts long-term electricity demand by the following formula: Among them, Y t It is represented by the long-term electricity demand forecast value at the current time t, c is a constant term, which represents the baseline level of the model, is the autoregressive part, which represents the relationship between the current value and the value at p time points in the past. p is the number of autoregressive terms, and φ i is the autoregressive coefficient, which indicates the influence of the lag period on the current value. is the moving average part, which represents the relationship between the current value and the random error of the past q time points, q is the number of moving average items, θ j is the moving average coefficient, which indicates the influence of the random error in the lag period on the current value, ε t is the random error term, indicating random fluctuations; The power demand forecast result is determined according to the short-term power demand forecast value and the long-term power demand forecast value.

6. The method according to claim 1, characterized in that The multiple types of data include meteorological data, operating data of power generation equipment, and historical power data. The power generation output is predicted by the second prediction model to obtain a power generation output prediction result, including: The second prediction model obtains the power generation output prediction result by the following formula: Among them, y i represents the power generation output prediction result output by the output layer, w hj represents the weight from the hidden layer to the output layer, σ represents the activation function, m represents the number of hidden layers, a h represents the input of the hidden layer neuron, w th represents the weight from the input layer to the hidden layer, x i represents the output of the input layer, b h represents the bias of the hidden layer, n represents the number of input layers, and the data input to the input layer at least includes the meteorological data, the operating status of the power generation equipment and the historical power data information.

7. The method according to claim 6, characterized in that The method further comprises: The output layer error is calculated by the following formula: Among them, e j represents the output layer error, represents the expected output of the output layer, y j Represents the actual output of the output layer; The hidden layer error is calculated by the following formula: Among them, δ h represents the error of the hidden layer, l represents the number of output layers, and σ' represents the derivative of the activation function; The weights from the input layer to the hidden layer are updated using the following formula: Among them, η represents the learning rate, represents the updated weight, Represents the weight before updating.

8. A device for determining a power generation strategy, characterized in that: include: An acquisition module, used to acquire different types of power generation strategy data; A prediction module, used to obtain a power demand prediction result output by a first prediction model based on the different types of power generation strategy data, and to obtain a power generation output prediction result output by a second prediction model based on the different types of power generation strategy data, wherein both the first prediction model and the second prediction model are trained based on historical power generation strategy data; A determination module is used to determine a target power generation strategy according to the power demand prediction result and the power generation output prediction result.

9. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to execute the method according to any one of claims 1 to 7 through the computer program.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images