Cooperative optimization method of multiple types of power sources based on complementary characteristics

Abstract

The present application provides a method for collaborative optimization of multiple types of power sources based on complementary characteristics, which relates to the field of power generation technology, including: calculating and obtaining a historical complementary coefficient sequence; obtaining a predicted renewable power generation parameter sequence; obtaining an optimal power generation control parameter; using the optimal power generation control parameter as a collaborative optimization result, and controlling the traditional power source to perform power generation control within a preset time range in the future. The present application can solve the technical problem that the existing technology relies too much on static or simple optimization models and fails to fully consider the impact of changes in meteorological data, resulting in poor complementary effects between traditional power sources and renewable power sources, thereby affecting the stability of power grid operation. By calculating the historical complementary coefficient sequence and performing meteorological data forecasting, combined with the prediction of renewable power generation parameters, the deviation between the complementary coefficient and the preset complementary coefficient is reduced, and the power generation control optimization of the traditional power source is realized, achieving the technical effect of improving power grid stability and energy utilization efficiency.

Description

Technical Field

[0001] The present application relates to the field of power generation technology, and in particular to a method for collaborative optimization of multiple types of power sources based on complementary characteristics. Background Art

[0002] With the continuous transformation of energy structure and the advancement of sustainable development goals, the proportion of renewable energy in modern power grids has gradually increased. Traditional power sources such as thermal power, hydropower and nuclear power have relatively stable output and long startup time, providing reliable support when dealing with load fluctuations. The volatility of renewable power sources requires the power grid to have flexible scheduling capabilities to ensure the stability of power supply. In this context, how to achieve the coordinated power generation of renewable and traditional power sources, make full use of the complementary characteristics of various power sources, and improve the efficiency and stability of power grid operation has become a key issue that needs to be solved urgently.

[0003] Existing multi-type power scheduling methods mostly rely on static or simple optimization models and fail to fully consider the impact of changes in meteorological data, resulting in poor complementarity between traditional power sources and renewable power sources, thus affecting the stability of power grid operation. Summary of the invention

[0004] The purpose of this application is to provide a method for collaborative optimization of multiple types of power sources based on complementary characteristics, in order to solve the technical problem that the prior art relies too much on static or simple optimization models and fails to fully consider the impact of changes in meteorological data, resulting in poor complementary effects between traditional power sources and renewable power sources, thereby affecting the stability of power grid operation.

[0005] In view of the above problems, the present application provides a method for collaborative optimization of multiple types of power sources based on complementary characteristics, including: collecting historical meteorological data sequences within a preset time range in the past, as well as historical renewable power generation parameter sequences and historical traditional power generation parameter sequences of the renewable power source and the traditional power source, and combining the load demand parameters of the power grid system to calculate and obtain a historical complementary coefficient sequence; based on the historical meteorological data sequence, forecasting meteorological data within a preset time range in the future to obtain a predicted meteorological data sequence, and forecasting the power generation parameters of the renewable power source to obtain a predicted renewable power generation parameter sequence; based on the predicted renewable power generation parameter sequence and the historical complementary coefficient sequence, optimizing the power generation control of the traditional power source within the preset time range in the future to obtain optimal power generation control parameters, wherein the optimization is performed with the purpose of reducing the complementary coefficient deviation from the preset complementary coefficient, wherein the complementary coefficient is the ratio of the sum of the renewable power generation parameters and the traditional power generation parameters to the load demand parameters; using the optimal power generation control parameters as the collaborative optimization result, within the preset time range in the future, controlling the traditional power source to perform power generation control.

[0006] The technical solution provided in this application has at least the following technical effects or advantages:

[0007] Collect historical meteorological data sequences within a preset time range in the past, as well as historical renewable power generation parameter sequences and historical traditional power generation parameter sequences of the renewable power source and the traditional power source, and combine them with the load demand parameters of the power grid system to calculate and obtain a historical complementary coefficient sequence; predict meteorological data within a preset time range in the future according to the historical meteorological data sequence to obtain a predicted meteorological data sequence, and predict the power generation parameters of the renewable power source to obtain a predicted renewable power generation parameter sequence; optimize the power generation control of the traditional power source within the preset time range in the future according to the predicted renewable power generation parameter sequence and the historical complementary coefficient sequence to obtain the optimal power generation control parameter, wherein the optimization is performed for the purpose of reducing the complementary coefficient deviation from the preset complementary coefficient, wherein the complementary coefficient is the ratio of the sum of the renewable power generation parameter and the traditional power generation parameter to the load demand parameter; use the optimal power generation control parameter as the collaborative optimization result, and control the traditional power source to perform power generation control within the preset time range in the future. By calculating the historical complementary coefficient sequence and performing meteorological data prediction, combined with the prediction of renewable power generation parameters, the deviation between the complementary coefficient and the preset complementary coefficient is reduced, and the power generation control optimization of the traditional power source is realized, so as to achieve the technical effect of improving the stability of the power grid and the efficiency of energy utilization.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0010] Figure 1 This is a flow chart of the collaborative optimization method of multiple types of power sources based on complementary characteristics in this application.

[0011] Figure 2This is a flow chart of obtaining optimal power generation control parameters in the collaborative optimization method of multiple types of power sources based on complementary characteristics in this application. DETAILED DESCRIPTION

[0012] This application solves the technical problem that the existing technology relies too much on static or simple optimization models and fails to fully consider the impact of changes in meteorological data, resulting in poor complementary effects between traditional power sources and renewable power sources, thus affecting the stability of power grid operation by providing a multi-type power source collaborative optimization method based on complementary characteristics. By calculating the historical complementary coefficient sequence and performing meteorological data forecasting, combined with the prediction of renewable power generation parameters, the deviation between the complementary coefficient and the preset complementary coefficient is reduced, the power generation control optimization of traditional power sources is realized, and the technical effect of improving power grid stability and energy utilization efficiency is achieved.

[0013] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0014] Please see attached Figure 1 The present application provides a method for collaborative optimization of multiple types of power sources based on complementary characteristics. The method for collaborative optimization of multiple types of power sources based on complementary characteristics is applied to a power grid system. The power grid system includes grid-connected renewable power sources and traditional power sources. The method for collaborative optimization of multiple types of power sources based on complementary characteristics specifically includes the following steps:

[0015] Step 1: Collect historical meteorological data sequences within a preset time range in the past, as well as historical renewable power generation parameter sequences and historical traditional power generation parameter sequences of the renewable power source and traditional power source, and calculate and obtain historical complementary coefficient sequences in combination with load demand parameters of the power grid system.

[0016] Specifically, in the power grid system, the coordination of renewable energy and traditional energy is an important task to ensure the stable operation of the power grid. The power grid system is an electric network composed of power generation, transmission, distribution and user power consumption, which is used to transmit and distribute electricity; grid-connected renewable power sources refer to power sources that are generated by renewable energy such as solar energy and wind energy and connected to the power grid, which are volatile and intermittent; traditional power sources refer to power sources that use fossil fuels (such as coal and natural gas) to generate electricity, which have stable power generation capacity and are usually used to balance the load of the power grid. The past preset time range refers to a period of time in the past, such as the past year.

[0017] The historical meteorological data sequence within the preset time range in the past, the historical renewable power generation parameter sequence of renewable power sources, and the historical traditional power generation parameter sequence of traditional power sources are collected. These data mainly cover the meteorological changes within the preset time range in the past and the power generation of the two types of power sources. At the same time, the load demand parameters of the power grid system must be obtained. The load demand parameters usually include the power consumption of the power grid in different time periods, reflecting the real-time demand of the power grid for electricity. By processing these data, the historical complementary coefficient sequence is calculated. The complementary coefficient essentially reflects the compensation ability of the two power sources to the load demand of the power grid. Specifically, the complementary coefficient is determined by the ratio of the sum of the power generation of renewable power sources and traditional power sources to the load demand of the power grid. Under ideal conditions, the complementary coefficient is close to 1, that is, the total power generation of the two power sources can just meet the load demand of the power grid and ensure the stable operation of the power grid. If the complementary coefficient is greater than 1, such as 1.1, it means that the power generation capacity of the power grid is slightly surplus, but it will not cause the power grid to be overloaded. Instead, it can provide a certain redundancy and improve the flexibility and emergency response capability of the system. If the complementary coefficient is too large or too small, it will cause an imbalance in the power grid. An excessively large complementary coefficient may lead to excess power generation in the power grid, affecting the effective use of electricity, while an excessively small complementary coefficient means that the load demand cannot be met and the power grid may face the risk of insufficient power supply. By calculating the historical complementary coefficient sequence, it provides a basis for subsequent power generation scheduling, thereby effectively reducing the power generation fluctuations caused by multiple types of power sources and improving the stability and reliability of the power grid.

[0018] Step 2: Based on the historical meteorological data sequence, forecast the meteorological data within a preset time range in the future to obtain a forecast meteorological data sequence, and forecast the power generation parameters of the renewable power source to obtain a forecast renewable power generation parameter sequence.

[0019] Specifically, by analyzing the changing trend of the historical meteorological data sequence, the existing machine learning model is used to predict the future meteorological data, and the predicted meteorological data sequence within the preset time range in the future is obtained. After obtaining the future predicted meteorological data sequence, the next step is to predict the power generation parameters of renewable power sources. The power generation of renewable power sources is closely related to meteorological conditions. Therefore, through regression analysis or machine learning modeling of historical meteorological data and historical renewable power generation data, a relationship model between meteorological data and power generation parameters can be established. For example, in wind power generation, there is an obvious correlation between wind speed and power generation; in photovoltaic power generation, light intensity and temperature have a direct impact on power generation. By analyzing the predicted meteorological data sequence, the predicted renewable power generation parameter sequence within the preset time range in the future can be predicted, which provides a basis for load prediction and optimal scheduling of the power grid system, and can provide an accurate basis for subsequent power coordination and power generation control optimization, reducing the instability caused by renewable energy fluctuations.

[0020] Step three: Based on the predicted renewable power generation parameter sequence and the historical complementary coefficient sequence, optimize the power generation control of the traditional power source within the future preset time range to obtain the optimal power generation control parameters, wherein the optimization is performed with the purpose of reducing the complementary coefficient deviation from the preset complementary coefficient, wherein the complementary coefficient is the ratio of the sum of the renewable power generation parameters and the traditional power generation parameters to the load demand parameters.

[0021] Specifically, the goal of optimizing power generation control for traditional power sources is to adjust the power generation parameters of traditional power sources to reduce the deviation of the complementary coefficient within the preset time range in the future, so that it is consistent with the preset complementary coefficient (such as 1). First of all, the power generation control parameters of traditional power sources usually include the output power, fuel consumption rate, power generation efficiency and other parameters of the generator set. By combining the predicted renewable power generation parameter sequence and optimizing the power generation control parameters of traditional power sources, the complementarity between renewable energy and traditional energy can be effectively achieved. In the optimization process, it is necessary to combine the predicted renewable power generation parameter sequence and the historical complementary coefficient sequence so that under the given load demand parameters and the predicted renewable power generation parameter sequence, the output power of the traditional power source can optimally match the load demand, while keeping the complementary coefficient as close to 1 as possible, so as to obtain the optimal power generation control parameters. The optimal power generation control parameters will be applied to the power generation control of traditional power sources to ensure that in the future time period, the power generation capacity of traditional power sources can be adjusted according to the changes of renewable power sources, so as to achieve a smooth supply of power grid load and keep the complementary coefficient consistent with the preset value, so as to ensure that the power grid system can operate smoothly when multiple types of power sources work together, avoid over-reliance on a certain type of power source, and thus improve the stability and efficiency of the power grid.

[0022] Step 4: Using the optimal power generation control parameters as the collaborative optimization result, control the traditional power source to perform power generation control within the future preset time range.

[0023] Specifically, after obtaining the optimal power generation control parameters, the next step is to apply them to the power generation control of traditional power sources to ensure the stable operation of the power grid and the balance of energy supply within a preset time range in the future, that is, to adjust the power generation of traditional power sources according to the optimal power generation control parameters, so as to effectively coordinate with the volatility and load demand of renewable energy, and ensure the smooth and efficient operation of the power grid.

[0024] Furthermore, step one of this application includes:

[0025] Collect meteorological data at multiple past moments within a preset time range in the past, as well as power generation parameters of the renewable power source and the traditional power source, to obtain a historical meteorological data sequence, a historical renewable power generation parameter sequence, and a historical traditional power generation parameter sequence; obtain load demand parameters of the power grid system; and calculate a historical complementary coefficient sequence of the renewable power source and the traditional power source based on the load demand parameters, the historical renewable power generation parameter sequence, and the historical traditional power generation parameter sequence.

[0026] Specifically, historical meteorological data includes factors that affect renewable energy generation, such as wind speed, light intensity, temperature, etc., while the power generation parameters of renewable power sources and traditional power sources reflect the power generation capacity of each power source in the power grid system under these meteorological conditions, such as power generation. The historical meteorological data sequence, the historical renewable power generation parameter sequence, and the historical traditional power generation parameter sequence are usually extracted based on the operation log of the existing power grid system. The meteorological data at multiple past moments and the power generation parameters of the renewable power source and the traditional power source are arranged in sequence to obtain the historical meteorological data sequence, the historical renewable power generation parameter sequence, and the historical traditional power generation parameter sequence.

[0027] At the same time, it is also necessary to obtain the load demand parameters of the power grid system. These data usually include the power demand of the power grid in different time periods. The load demand parameters reflect the power demand intensity of the power grid at each moment. They usually come from the real-time monitoring system of the power grid dispatching center. The load demand parameters can be combined with historical meteorological data and power generation parameters to calculate the historical complementary coefficient. The complementary coefficient measures the degree of coordination between renewable power sources and traditional power sources in meeting the load demand of the power grid. The specific calculation method is to add the power generation parameters of renewable power sources and traditional power sources in the historical moment, and then calculate the ratio with the load demand parameters of the power grid. Assuming that the load demand of the power grid at a certain moment is 600MW, the power generation of renewable power sources is 350MW, and the power generation of traditional power sources is 250MW, then the complementary coefficient at this moment is (350+250) / 600=1.0. A complementary coefficient of 1.0 means that the two just meet the load demand of the power grid; if the coefficient is greater than 1.0, it means that the power supply capacity of the power grid exceeds the demand, otherwise there may be a risk of insufficient power supply.

[0028] By combining load demand parameters, historical renewable power generation parameter sequences and historical traditional power generation parameter sequences, the complementary coefficients at each moment can be calculated and formed into a historical complementary coefficient sequence, providing basic data for subsequent power generation control optimization and helping to optimize future power grid dispatching strategies.

[0029] Furthermore, the present application also includes the following steps:

[0030] According to the historical renewable power generation parameter sequence and the historical traditional power generation parameter sequence, the sum of multiple past power generation parameters at the multiple past moments is calculated; the ratio of the sum of the multiple past power generation parameters to the load demand parameter is calculated to obtain multiple historical complementary coefficients, and a historical complementary coefficient sequence is constructed in combination with the multiple past moments.

[0031] Specifically, according to the historical renewable power generation parameter sequence and the historical traditional power generation parameter sequence, the renewable energy power generation (such as wind power, solar power, etc.) and the traditional power generation (such as coal power, nuclear power, etc.) at each moment are added to obtain the total power generation at each moment, that is, the sum of multiple past power generation parameters. Next, the ratio of the sum of multiple past power generation parameters to the grid load demand parameter is calculated to obtain multiple historical complementary coefficients. For example, assuming that the load demand of the grid at a certain moment is 650MW, and the corresponding total power generation is 700MW, then the complementary coefficient at that moment is 700MW / 650MW=1.08.

[0032] Finally, multiple historical complementary coefficients are arranged in chronological order to form a historical complementary coefficient sequence, which reflects the matching between the power supply capacity and demand of the power grid at various moments in history. In practical applications, this sequence is the basis for future power grid dispatch optimization. The analysis of historical complementary coefficients can help predict possible supply and demand fluctuations in the future, thereby making more accurate power dispatch and optimization decisions.

[0033] Further, step 2 of this application includes:

[0034] According to the meteorological monitoring data in the historical time, a set of sample historical meteorological data sequences is collected, and a future meteorological data sequence within a preset time range after each sample historical meteorological data sequence is collected to obtain a set of sample future meteorological data sequences; the sample historical meteorological data sequence set and the sample future meteorological data sequence set are used to train a meteorological data prediction channel; the historical meteorological data sequence is input into the meteorological data prediction channel, and the prediction output is used to obtain a predicted meteorological data sequence within the preset time range in the future.

[0035] Specifically, in order to predict future meteorological data, it is first necessary to collect historical meteorological monitoring data, which is usually provided by meteorological stations and sensor networks, including variables such as temperature, wind speed, humidity, and radiation intensity, which are usually collected in units of hours or days, thereby obtaining a set of sample historical meteorological data sequences, which includes multiple sample historical meteorological data sequences. Next, after each sample historical meteorological data sequence, future meteorological data sequences within a preset time range after that time point are collected. For example, based on each sample historical meteorological data sequence, meteorological data corresponding to the next 24 hours are collected to form a corresponding set of future meteorological data sequences.

[0036] Furthermore, the sample historical meteorological data sequence set and the sample future meteorological data sequence set are trained using machine learning or deep learning methods to train the meteorological data prediction channel. Common models include time series prediction models, neural network-based models (such as long short-term memory neural networks), etc. The training process is supervised learning through the historical meteorological data sequence and the corresponding sample future meteorological data sequence, so that the meteorological data prediction channel can identify the pattern in the historical data and accurately infer the future meteorological change trend in the prediction process. Taking LSTM (Long Short-Term Memory Neural Network) as an example, the structure of the meteorological data prediction channel includes: input layer: convert the sample historical meteorological data sequence into a time step format and input it into the model. The data of each time step includes different meteorological variables (such as temperature, wind speed, air pressure, etc.); LSTM layer: contains multiple LSTM units, which can capture the long-term time dependency in the input data. For example, the LSTM unit will learn the weather change pattern of the past week and predict the meteorological data of the next week; output layer: the LSTM network outputs the meteorological data of the future time period (such as the temperature and humidity of the next 24 hours), which will be used as the meteorological prediction result.

[0037] After completing the model training, the historical meteorological data sequence is input into the trained meteorological data prediction channel to predict future meteorological data and obtain a predicted meteorological data sequence. These predicted data will be used for the next step of power generation prediction, load prediction and other operations to optimize the energy scheduling and configuration of the power grid.

[0038] Furthermore, step 2 of this application also includes:

[0039] According to the renewable power generation monitoring data in historical time, a sample meteorological data set is collected, and the power generation parameters of the renewable power source under different sample meteorological data are collected to obtain a sample renewable power generation parameter set; the sample meteorological data set and the sample renewable power generation parameter set are used to train a renewable power generation prediction channel; multiple predicted meteorological data in the predicted meteorological data sequence are input into the renewable power generation prediction channel, and multiple renewable power generation parameters are obtained by prediction output, and a predicted renewable power generation parameter sequence is constructed in combination with multiple future moments within the future preset time range.

[0040] Specifically, historical data in the past time (a period of time in the past, such as the past year) is used for sample collection. By collecting historical meteorological data and the power generation monitoring data of renewable power sources in the corresponding time period, a sample meteorological data set and a sample renewable power generation parameter set are formed. Each sample meteorological data contains meteorological data (such as wind speed, radiation intensity, temperature, humidity, etc.), and the sample renewable power generation parameters are the renewable power generation data at the corresponding moment (such as wind power generation, solar power generation, etc.). Next, the renewable power generation prediction channel is trained using existing machine learning models (such as regression analysis, neural networks or support vector machines). The input of the renewable power generation prediction channel is meteorological data, and the output is the power generation parameters of the corresponding renewable power source. For example, based on historical wind speed and radiation data, the renewable power generation prediction channel can learn how to predict the power generation capacity of wind turbines and photovoltaic cells.

[0041] After completing the training of the renewable power generation prediction channel, the renewable power generation prediction channel is used to predict future meteorological data within a future preset time range. Specifically, multiple predicted meteorological data in the predicted meteorological data sequence are input into the trained renewable power generation prediction channel, and the renewable power generation prediction channel processes the input meteorological data to predict the renewable power generation parameters corresponding to multiple future moments within the future preset time range. Finally, a complete predicted renewable power generation parameter sequence is constructed in combination with multiple predicted moments within the future preset time range. This sequence can provide the power grid dispatching system with information on the power generation capacity of renewable power sources in the future time period, thereby providing a basis for power grid load forecasting, traditional power supply dispatching and power grid optimization.

[0042] Further, as attached Figure 2 As shown, step three of this application includes:

[0043] Acquire the power generation control parameter space of the traditional power source; randomly generate a first power generation control parameter in the power generation control parameter space; predict the power generation parameters of the traditional power source for the first power generation control parameter according to the predicted meteorological data sequence, and obtain a predicted traditional power generation parameter sequence; calculate and obtain a predicted complementary coefficient sequence based on the predicted renewable power generation parameter sequence and the predicted traditional power generation parameter sequence and the load demand parameter, and calculate and obtain a first synergistic fitness by combining the historical complementary coefficient sequence; continue to optimize the power generation control parameters of the traditional power source until convergence, output the power generation control parameters with the maximum synergistic fitness, and obtain the optimal power generation control parameters.

[0044] Specifically, the power generation control parameter space of the traditional power source includes all possible power generation control parameter combinations, such as power generation, fuel consumption rate, power generation efficiency, etc. These control parameters determine the output capacity of the traditional power source. In order to optimize these parameters, an initial power generation control parameter is first randomly generated in the power generation control parameter space as the first power generation control parameter. Next, according to the predicted meteorological data sequence, the first power generation control parameter is used to predict the power generation parameters of the traditional power source. The power generation parameter prediction usually uses a machine learning model such as regression analysis, machine learning algorithm or deep neural network to obtain a predicted traditional power generation parameter sequence. Then, based on the load demand parameter, combined with the predicted renewable power generation parameter sequence and the predicted traditional power generation parameter sequence, the predicted renewable power generation parameters and the predicted traditional power generation parameters at the same time are added, and the ratio of the addition result to the load demand parameter is calculated to obtain a predicted complementary coefficient sequence.

[0045] In order to further optimize the power generation control parameters of the traditional power source, the historical complementary coefficient sequence is used as a reference to calculate the first synergistic fitness of the first power generation control parameter. The synergistic fitness is an indicator to measure the synergistic effect between the traditional power source and the renewable power source. It is usually defined as the difference between the historical complementary coefficient sequence and the predicted complementary coefficient sequence. The greater the synergistic fitness, the better the selected power generation control parameter can achieve the stability of the power grid and the balance of load demand. Next, the power generation control parameters of the traditional power source are adjusted through optimization algorithms (such as genetic algorithms, particle swarm optimization, gradient descent, etc.). After each adjustment, a new traditional power generation parameter sequence is calculated based on the new control parameters, and the predicted complementary coefficient sequence and its synergistic fitness are updated. This process will be repeated many times until the synergistic fitness converges. The condition for judging convergence can be the number of iterations, such as 500 iterations. Finally, the power generation control parameter with the maximum synergistic fitness is output as the optimal power generation control parameter. Through this optimization method, the output of the traditional power source can be adjusted in real time to maximize the use of renewable energy while ensuring the stability of the power grid.

[0046] Furthermore, the present application also includes the following steps:

[0047] According to the power generation data of the traditional power source in the historical time, a sample power generation control parameter set and a sample meteorological data set are collected, and the traditional power generation parameters of different sample power generation control parameters under the influence of different sample meteorological data are collected to obtain a sample traditional power generation parameter set; the sample power generation control parameter set, the sample meteorological data set and the sample traditional power generation parameter set are used to train a traditional power generation prediction channel; multiple predicted meteorological data at multiple future moments in the predicted meteorological data sequence are input into the traditional power generation prediction channel in combination with the first power generation control parameter, and the predicted output obtains the predicted traditional power generation parameters at the multiple future moments to obtain a predicted traditional power generation parameter sequence.

[0048] Specifically, according to the predicted meteorological data sequence, the first power generation control parameter is used to predict the power generation parameters of the traditional power source, and the specific steps of obtaining the predicted traditional power generation parameter sequence include: the historical power generation data of the traditional power source generally includes its power generation, power output, meteorological data during power generation, etc. at different times and under different meteorological conditions, based on which a sample power generation control parameter set and a sample meteorological data set are collected, and the data in the sample power generation control parameter set and the sample meteorological data set have a one-to-one correspondence. The sample power generation control parameter set contains parameters related to power generation control, such as combustion rate, generator output power, fuel consumption, etc.; and the sample meteorological data set includes historical meteorological data, such as temperature, humidity, wind speed, sunshine intensity, etc., which meteorological factors affect the power generation capacity of the traditional power source. The sample traditional power generation parameter set is the traditional power generation parameter of different sample power generation control parameters under the influence of different sample meteorological data corresponding to the sample power generation control parameter set and the sample meteorological data set, such as power generation. That is to say, the power generation of traditional power sources will also be affected by meteorological data, such as the temperature affecting power generation, so it is also necessary to combine meteorological data for prediction.

[0049] Further, the sample power generation control parameter set, the sample meteorological data set and the sample traditional power generation parameter set are used to train the traditional power generation prediction channel. The traditional power generation prediction channel usually uses machine learning or deep learning algorithms to establish models, such as support vector machines (SVM), random forests, neural networks, etc. These models learn the complex relationship between input and output based on sample power generation control parameters, meteorological data and historical power generation parameters, so that future power generation parameters can be predicted based on new meteorological conditions and control parameters. That is, the input of the traditional power generation prediction channel is power generation control parameters and meteorological data, and the output is traditional power generation parameters. Exemplarily, the sample power generation control parameter set, the sample meteorological data set and the sample traditional power generation parameter set are divided into a training set and a validation set, and the model is trained with the training set, and the learning parameters are optimized by minimizing the loss function (such as mean square error). During the training process, the validation set is used to evaluate the generalization ability of the traditional power generation prediction channel to prevent overfitting. After the training is completed, the model is evaluated using an independent test set to ensure that the prediction ability of the traditional power generation prediction channel is good.

[0050] After combining multiple predicted meteorological data at multiple future moments in the predicted meteorological data sequence with the first power generation control parameter, the data are input into the traditional power generation prediction channel to perform power generation parameter prediction to obtain the traditional power generation parameters at each future moment. These predicted values ​​constitute a predicted traditional power generation parameter sequence of the traditional power source in a future period of time, providing data support for subsequent optimization decisions, facilitating adjustment of the output of the traditional power source, and ensuring that the power grid can operate smoothly in the future time period to meet the load demand.

[0051] Furthermore, the present application also includes the following steps:

[0052] According to the predicted renewable power generation parameter sequence and the predicted traditional power generation parameter sequence, the sum of multiple future power generation parameters at multiple future moments is calculated; the ratio of the sum of the multiple future power generation parameters to the load demand parameter is respectively calculated to obtain multiple predicted complementary coefficients, and a predicted complementary coefficient sequence is constructed; according to the historical complementary coefficient sequence and the predicted complementary coefficient sequence, the first coordinated fitness of the first power generation control parameter is calculated, as shown in the following formula: ; Among them, XTA is the first collaborative fitness, and is the weight, M is the number of historical complementary coefficients and predicted complementary coefficients in the historical complementary coefficient sequence and the predicted complementary coefficient sequence, is the i-th complementary coefficient, To preset complementary coefficients, is the mean of the complementary coefficients of the historical complementary coefficient sequence and the predicted complementary coefficient sequence.

[0053] Specifically, according to the predicted renewable power generation parameter sequence and the predicted traditional power generation parameter sequence, the predicted renewable power generation parameters and the predicted traditional power generation parameters at multiple moments in the future are added to obtain the sum of multiple future power generation parameters. Next, the ratio of the sum of future power generation parameters at each future moment to the load demand parameter is calculated to obtain multiple predicted complementary coefficients, and the coefficients are serialized in chronological order to obtain a predicted complementary coefficient sequence. If the complementary coefficient is close to 1, it means that the power generation and load demand are well matched; if the complementary coefficient is greater than 1, it means that the power generation exceeds the demand, and if it is less than 1, it means that the power generation is insufficient. The calculation formula for determining the first collaborative fitness is as follows: ; Among them, XTA is the first collaborative fitness, and is the weight, M is the number of historical complementary coefficients and predicted complementary coefficients in the historical complementary coefficient sequence and the predicted complementary coefficient sequence, is the i-th complementary coefficient, To preset complementary coefficients, is the mean of the complementary coefficients of the historical complementary coefficient sequence and the predicted complementary coefficient sequence.

[0054] The calculation formula of the first collaborative fitness includes two items. The first item reduces the deviation between the complementary coefficient and the preset complementary coefficient. The smaller the difference, the greater the fitness. The second item reduces the variance of all complementary coefficients. The smaller the variance, the more stable the power generation complementarity and the greater the fitness. The preset complementary coefficient is set by the technicians in this field, for example, 1, or slightly greater than 1 (for example, 1.1). In simple terms, the historical complementary coefficient sequence is the complementary coefficient within the preset time range in the past, and the predicted complementary coefficient sequence is the predicted complementary coefficient within the preset time range in the future. The historical complementary coefficient sequence is fixed, but the predicted complementary coefficient sequence will change with the iterative optimization of the power generation control parameters of the traditional power source. Substituting all the complementary coefficients in the historical complementary coefficient sequence and the predicted complementary coefficient sequence into the formula, the first collaborative fitness can be calculated, so as to find the power generation control parameters that maximize the collaborative fitness and ensure that the power grid system can operate efficiently and stably in the future time range.

[0055] In summary, the multi-type power supply collaborative optimization method based on complementary characteristics provided by the present application has the following technical effects:

[0056] Collect historical meteorological data sequences within a preset time range in the past, as well as historical renewable power generation parameter sequences and historical traditional power generation parameter sequences of the renewable power source and the traditional power source, and combine them with the load demand parameters of the power grid system to calculate and obtain a historical complementary coefficient sequence; predict meteorological data within a preset time range in the future according to the historical meteorological data sequence to obtain a predicted meteorological data sequence, and predict the power generation parameters of the renewable power source to obtain a predicted renewable power generation parameter sequence; optimize the power generation control of the traditional power source within the preset time range in the future according to the predicted renewable power generation parameter sequence and the historical complementary coefficient sequence to obtain the optimal power generation control parameter, wherein the optimization is performed for the purpose of reducing the complementary coefficient deviation from the preset complementary coefficient, wherein the complementary coefficient is the ratio of the sum of the renewable power generation parameter and the traditional power generation parameter to the load demand parameter; use the optimal power generation control parameter as the collaborative optimization result, and control the traditional power source to perform power generation control within the preset time range in the future. By calculating the historical complementary coefficient sequence and performing meteorological data prediction, combined with the prediction of renewable power generation parameters, the deviation between the complementary coefficient and the preset complementary coefficient is reduced, and the power generation control optimization of the traditional power source is realized, so as to achieve the technical effect of improving the stability of the power grid and the efficiency of energy utilization.

[0057] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0058] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A multi-type power supply collaborative optimization method based on complementary characteristics, characterized in that: The method for collaborative optimization of multiple types of power sources based on complementary characteristics is applied to a power grid system, wherein the power grid system includes grid-connected renewable power sources and traditional power sources, and the method for collaborative optimization of multiple types of power sources based on complementary characteristics includes: Collecting a historical meteorological data sequence within a preset time range in the past, as well as a historical renewable power generation parameter sequence and a historical traditional power generation parameter sequence of the renewable power source and the traditional power source, and combining the load demand parameters of the power grid system to calculate and obtain a historical complementary coefficient sequence; According to the historical meteorological data sequence, forecasting meteorological data within a preset time range in the future is performed to obtain a forecasted meteorological data sequence, and forecasting power generation parameters of the renewable power source is performed to obtain a forecasted renewable power generation parameter sequence; According to the predicted renewable power generation parameter sequence and the historical complementary coefficient sequence, the power generation control optimization of the traditional power source within the future preset time range is performed to obtain the optimal power generation control parameter, wherein the optimization is performed with the purpose of reducing the complementary coefficient deviation between the complementary coefficient and the preset complementary coefficient, wherein the complementary coefficient is the ratio of the sum of the renewable power generation parameter and the traditional power generation parameter to the load demand parameter; Using the optimal power generation control parameter as a collaborative optimization result, controlling the conventional power source to perform power generation control within the future preset time range; Collecting a historical meteorological data sequence within a preset time range in the past, as well as a historical renewable power generation parameter sequence and a historical traditional power generation parameter sequence of the renewable power source and the traditional power source, and combining the load demand parameters of the power grid system to calculate and obtain a historical complementary coefficient sequence, including: Collect meteorological data at multiple past moments within a preset time range, and power generation parameters of the renewable power source and the traditional power source, to obtain a historical meteorological data sequence, a historical renewable power generation parameter sequence, and a historical traditional power generation parameter sequence; Obtaining load demand parameters of the power grid system; According to the load demand parameter, the historical renewable power generation parameter sequence and the historical traditional power generation parameter sequence, the historical complementary coefficient sequence of the renewable power source and the traditional power source is calculated.

2. The method for collaborative optimization of multiple types of power sources based on complementary characteristics according to claim 1, characterized in that: According to the load demand parameter, the historical renewable power generation parameter sequence and the historical traditional power generation parameter sequence, the historical complementary coefficient sequence of the renewable power source and the traditional power source is calculated, including: According to the historical renewable power generation parameter sequence and the historical conventional power generation parameter sequence, a sum of the multiple past power generation parameters at the multiple past moments is calculated; The ratio of the sum of the multiple past power generation parameters to the load demand parameter is calculated to obtain multiple historical complementary coefficients, and a historical complementary coefficient sequence is constructed in combination with the multiple past moments.

3. The method for collaborative optimization of multiple types of power sources based on complementary characteristics according to claim 1, characterized in that: Based on the historical meteorological data sequence, the meteorological data within a preset time range in the future is predicted, including: According to the meteorological monitoring data in the historical time, a set of sample historical meteorological data sequences is collected, and a future meteorological data sequence within a preset time range after each sample historical meteorological data sequence is collected to obtain a set of sample future meteorological data sequences; Using the sample historical meteorological data sequence set and the sample future meteorological data sequence set to train a meteorological data prediction channel; The historical meteorological data sequence is input into the meteorological data prediction channel, and the prediction output obtains the predicted meteorological data sequence within the future preset time range.

4. The method for collaborative optimization of multiple types of power sources based on complementary characteristics according to claim 1, characterized in that: Predicting the power generation parameters of the renewable power source to obtain a predicted renewable power generation parameter sequence includes: According to the monitoring data of renewable power generation in historical time, a set of sample meteorological data is collected, and the power generation parameters of the renewable power source under different sample meteorological data are collected to obtain a set of sample renewable power generation parameters; Using the sample meteorological data set and the sample renewable power generation parameter set to train a renewable power generation prediction channel; The multiple predicted meteorological data in the predicted meteorological data sequence are input into the renewable power generation prediction channel, and the prediction output obtains multiple renewable power generation parameters. Combined with multiple future moments within the future preset time range, a predicted renewable power generation parameter sequence is constructed.

5. The method for collaborative optimization of multiple types of power sources based on complementary characteristics according to claim 1, characterized in that: According to the predicted renewable power generation parameter sequence and the historical complementary coefficient sequence, the power generation control optimization of the conventional power source within the future preset time range is performed to obtain the optimal power generation control parameters, including: Acquiring a power generation control parameter space of the conventional power source; Randomly generating a first power generation control parameter in the power generation control parameter space; According to the predicted meteorological data sequence, predicting the power generation parameters of the conventional power source for the first power generation control parameter to obtain a predicted conventional power generation parameter sequence; According to the predicted renewable power generation parameter sequence and the predicted traditional power generation parameter sequence, based on the load demand parameter, a predicted complementary coefficient sequence is calculated, and combined with the historical complementary coefficient sequence, a first collaborative fitness is calculated; The power generation control parameters of the conventional power source are continuously optimized until convergence, and the power generation control parameters with the greatest collaborative fitness are output to obtain the optimal power generation control parameters.

6. The method for collaborative optimization of multiple types of power sources based on complementary characteristics according to claim 5 is characterized in that: According to the predicted meteorological data sequence, predicting the power generation parameters of the conventional power source for the first power generation control parameter to obtain a predicted conventional power generation parameter sequence includes: According to the power generation data of the traditional power source in the historical period, a sample power generation control parameter set and a sample meteorological data set are collected, and the power generation parameters of the traditional power source with different sample power generation control parameters under the influence of different sample meteorological data are collected to obtain a sample traditional power generation parameter set; Using the sample power generation control parameter set, the sample meteorological data set and the sample traditional power generation parameter set to train a traditional power generation prediction channel; The multiple predicted meteorological data at multiple future moments in the predicted meteorological data sequence are combined with the first power generation control parameter and input into the traditional power generation prediction channel, and the predicted traditional power generation parameters at the multiple future moments are obtained through prediction output to obtain a predicted traditional power generation parameter sequence.

7. The method for collaborative optimization of multiple types of power sources based on complementary characteristics according to claim 5, characterized in that: According to the predicted renewable power generation parameter sequence and the predicted traditional power generation parameter sequence, based on the load demand parameter, a predicted complementary coefficient sequence is calculated, and combined with the historical complementary coefficient sequence, a first collaborative fitness is calculated, including: According to the predicted renewable power generation parameter sequence and the predicted traditional power generation parameter sequence, a sum of multiple future power generation parameters at multiple future moments is calculated; Respectively calculating the ratio of the sum of the multiple future power generation parameters to the load demand parameter, obtaining multiple prediction complementary coefficients, and constructing a prediction complementary coefficient sequence; According to the historical complementary coefficient sequence and the predicted complementary coefficient sequence, the first coordinated fitness of the first power generation control parameter is calculated and obtained as follows: ； Among them, XTA is the first collaborative fitness, and is the weight, M is the number of historical complementary coefficients and predicted complementary coefficients in the historical complementary coefficient sequence and the predicted complementary coefficient sequence, is the i-th complementary coefficient, To preset complementary coefficients, is the mean of the complementary coefficients of the historical complementary coefficient sequence and the predicted complementary coefficient sequence.

Images