A steam energy storage device group control method

Abstract

The present invention discloses a method and equipment for regulating a group of steam energy storage devices, and relates to the field of steam energy storage device groups. The thermal load prediction model of the present invention uses a BP neural network prediction algorithm to predict load demand. The control strategy can predict the optimal power output based on historical data and real-time data. The control method is applied to the steam energy storage device group to achieve coordinated control of the steam energy storage device group, and improve the accuracy and adaptability of the control strategy. At the same time, according to the load demand prediction, the steam demand of the target load can be predicted and calculated by comparing the steam demand of the terminal user. The use of renewable energy power during the day cannot meet the sufficient steam demand. During the off-peak period of night, low electricity prices can be used to prepare supplementary steam and store the steam, so as to fully reduce the steam cost and make up for the shortage of steam of renewable energy power, and finally realize the conversion of unstable energy of renewable energy power into stable steam output.

Description

Technical Field

[0001] The present invention relates to the field of steam energy storage device groups, and in particular to a method for regulating and controlling steam energy storage device groups. Background Art

[0002] Energy is an important driving force for world development. Human society has long used energy mainly based on fossil fuels, which has caused a large amount of greenhouse gas emissions and brought about serious global warming problems. With the increasing requirements of the country for environmental pollution control, large-scale reduction and gradual elimination of fossil energy has become a general trend. With China's large-scale development of new energy technologies such as wind power generation, the production cost per kilowatt-hour of wind power generation in many areas with good natural conditions is already lower than the comprehensive cost of traditional thermal power generation. This makes it technically possible to replace fossil energy with green zero-carbon electricity and produce low-cost green zero-carbon steam in the future, and it is also gradually becoming economical.

[0003] In the prior art, such as the Chinese publication number CN112907118A, the invention provides an energy distribution system for off-peak electricity steam accumulator, including: multiple user terminals, regional management terminals and remote terminals, wherein the user terminal is used to collect the user's energy consumption information; the regional management terminal is used to send the user's energy consumption information to the remote terminal; the remote terminal is used to convert and store energy according to the user's energy consumption information during off-peak electricity, and distribute the converted stored energy to the regional management terminal; the regional management terminal is also used to distribute the energy distributed by the remote terminal to multiple user terminals according to the user's energy consumption information, for use by multiple user terminals. In the system of the present invention, the remote terminal converts and stores energy according to the user's energy consumption information during off-peak electricity, and distributes the converted energy to the regional management terminal. When the user terminal uses energy, the regional management terminal distributes energy according to the energy consumption information of the user at the user terminal, so as to store as much as needed, thereby avoiding energy waste to the greatest extent.

[0004] Steam energy storage equipment is used to store and release steam. In order to meet the needs of industrial enterprises for high-pressure steam production, steam energy storage equipment requires stable power input to supply stable steam energy. Some directly purchase electricity through the urban power grid to supply steam thermal energy. Although this method can stably provide the required amount of high-pressure steam, it does not utilize renewable power generation energy such as wind power generation, which does not meet the national requirements for environmental pollution control and also increases industrial costs. Some directly use renewable power generation energy to supply electricity, but the supply of renewable power generation energy is sometimes not very stable. For example, wind power generation generates less power when the wind is low, and photovoltaic power generation cannot generate electricity at night, which further cannot meet the needs of industrial production steam. Therefore, it is necessary to convert unstable energy into stable steam input to achieve stable industrial steam transportation. If the steam energy storage equipment group is reasonably regulated, energy utilization efficiency can be improved. Summary of the invention

[0005] The purpose of the present invention is to provide a steam energy storage device group control method for reasonably regulating the steam energy storage device group, improving energy utilization efficiency, converting unstable energy into stable steam input, achieving stable transportation of industrial steam, and meeting the needs of the factory.

[0006] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0007] A method for controlling a group of steam energy storage devices comprises the following steps:

[0008] S1. Transmit unstable, unstable and discontinuous renewable energy electricity to a group of steam energy storage devices;

[0009] S2. The thermal load prediction model combines the energy storage and release data of the steam energy storage device group to predict the load demand;

[0010] S3, generating a control strategy through a heat load prediction model while performing distribution network control and optimization strategies to achieve the best operating state of the steam energy storage equipment;

[0011] S4. Control strategy applies the control method to the steam energy storage device group to achieve coordinated control of the steam energy storage device group.

[0012] Preferably, in step S1, the unstable, unstable and non-continuous renewable new energy electricity includes solar photovoltaic power generation, wind power generation and power abandonment during grid off-peak hours, wherein the power abandonment during grid off-peak hours refers to the power abandoned when the grid operator shuts down or reduces production of some power generation equipment in order to maintain the balance and stability of the grid in the case of oversupply of electricity, resulting in part of the electricity being unable to be used, and the hydropower abandonment refers to the power abandoned when the hydropower station cannot convert all the water energy into electricity due to electricity demand when generating electricity.

[0013] Preferably, in step S1, the steam energy storage device group includes a control system and a control device, wherein the control system includes a sensor module, a model training module, a control strategy generation module and a control execution module, and the control device includes multiple steam energy storage devices, a sensor network and a control center.

[0014] Preferably, in step S1, the sensor module is used to obtain historical data and real-time data related to the steam energy storage device group, the model training module is used to train and optimize the thermal load prediction model, and the control model is trained based on the historical data and real-time data. The control strategy generation module is used to generate a control strategy according to the control model, including adjusting the power output of each device in the group to achieve load balance of the group, and the control execution module is used to apply the control strategy to the steam energy storage device group to achieve coordinated control of the group. The multiple steam energy storage devices are used to store and release steam energy. The sensor network includes temperature sensors, pressure sensors and flow sensors, which are used to obtain real-time data related to the steam energy storage device group. The control center is used to train the control model, generate the control strategy and apply it to the steam energy storage device group.

[0015] Preferably, in step S2, the heat load prediction model predicts the load demand based on a BP neural network prediction algorithm, and simultaneously uses a genetic algorithm to optimize the BP neural network prediction algorithm.

[0016] Preferably, in step S2, the BP neural network prediction algorithm learns and trains the input sample data, mines the rules hidden in the data, and uses the learned rules to predict future load data. The genetic algorithm uses a parallel random search optimization method to select individuals according to the fitness function to retain individuals with strong fitness and eliminate individuals with poor fitness, while the newly formed group inherits the excellent information of the previous generation. The genetic algorithm is a global optimization algorithm used to make up for the shortcomings of the BP neural network algorithm falling into a local minimum, screen out the optimal BP weights and thresholds, and improve prediction accuracy and convergence speed.

[0017] Preferably, in step S2, the BP neural network performs load demand prediction including the following sub-steps:

[0018] S21, selecting the temperature, pressure, flow data, power generation and heat load influencing factors of the steam energy storage equipment group for the previous three days as input variables;

[0019] S22, preprocessing the data and taking the predicted daily heat load as the output variable;

[0020] S23, determining the structure of the BP neural network, including the number of nodes in the input layer, the hidden layer, and the output layer and their connection mode, and using a smaller random value for initialization to set the parameters of the genetic algorithm;

[0021] S24, calculating the fitness by the training error of the BP neural network, and performing crossover and mutation operations when conditions are met;

[0022] S25. When the conditions are not met, the optimal weighted network value is obtained, the error is measured, the weighted threshold is updated, and the load demand is predicted when the conditions are met.

[0023] Preferably, in step S23, the hidden layer is one layer, and the number of hidden layer nodes is , where π is the number of input variables.

[0024] Preferably, in step S22, the preprocessing includes denoising, filtering and normalizing the data to improve the accuracy and usability of the data.

[0025] Preferably, in step S3, the control strategy includes adjusting the power output of each device in the group to achieve the optimal operating state of the steam energy storage device. At the same time, through the imitation learning mechanism, the control strategy can predict the optimal power output based on historical data and real-time data to meet the needs of the power grid and maximize energy utilization efficiency.

[0026] The beneficial effects of the present invention are:

[0027] The present invention transmits unstable, unstable and non-continuous renewable new energy electricity to a group of steam energy storage devices. The thermal load prediction model uses a BP neural network prediction algorithm and is optimized by a genetic algorithm to predict the load demand. At this time, the control strategy can predict the optimal power output based on historical data and real-time data to meet the needs of the power grid and maximize energy utilization efficiency. The control method is applied to the steam energy storage device group to achieve coordinated control of the steam energy storage device group, improve the accuracy and adaptability of the control strategy, and at the same time, based on the load demand prediction, the steam demand of the terminal user can be compared to predict and calculate the target load steam demand. When the difference in user steam demand cannot be met, the use of renewable new energy electricity during the day cannot meet the sufficient steam demand. During the off-peak electricity period at night, low electricity prices can be used to prepare supplementary steam and store the steam to fully reduce the steam cost and make up for the shortage of steam of renewable new energy electricity, and finally realize the conversion of unstable energy of renewable energy electricity into stable steam input. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a flow chart of the steps of an embodiment of a method for controlling a group of steam energy storage devices of the present invention.

[0029] Figure 2 The present invention is a flow chart of the steps of predicting load demand using a BP neural network in a method for controlling a group of steam energy storage devices.

[0030] Figure 3 The present invention is a flow chart of the control system steps in a method for controlling a group of steam energy storage devices. DETAILED DESCRIPTION

[0031] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the present invention is further described below in conjunction with specific embodiments, but the following embodiments are only preferred embodiments of the present invention, not all. Based on the embodiments in the implementation mode, other embodiments obtained by those skilled in the art without creative work all belong to the protection scope of the present invention. The experimental methods in the following embodiments, unless otherwise specified, are conventional methods, and the materials, reagents, etc. used in the following embodiments, unless otherwise specified, can all be obtained from commercial channels.

[0032] Combine the following Figure 1 The present invention is further described. Example

[0033] A method for controlling a group of steam energy storage devices comprises the following steps:

[0034] S1. Transmit unstable, unstable and discontinuous renewable energy electricity to a group of steam energy storage devices.

[0035] The present invention is further configured as follows: unstable, non-constant and non-sustainable renewable new energy electricity includes solar photovoltaic power generation, wind power generation and power abandonment during grid off-peak hours, wherein power abandonment during grid off-peak hours refers to the situation where there is an oversupply of electricity and the grid operator shuts down or reduces production of some power generation equipment to maintain the balance and stability of the grid, resulting in a portion of the electricity being unable to be utilized and abandoned; hydropower abandonment refers to the electricity abandoned when a hydropower station is generating electricity and cannot convert all water energy into electricity due to electricity demand.

[0036] The present invention is further configured as follows: the steam energy storage device group includes a control system and a control device, wherein the control system includes a sensor module, a model training module, a control strategy generation module and a control execution module, and the control device includes multiple steam energy storage devices, a sensor network and a control center.

[0037] The present invention is further configured as follows: a sensor module is used to obtain historical data and real-time data related to a group of steam energy storage devices; a model training module is used to train and optimize a heat load prediction model; a control model is trained based on historical data and real-time data; a control strategy generation module is used to generate a control strategy according to the control model, including adjusting the power output of each device in the group to achieve load balance of the group; a control execution module is used to apply the control strategy to the steam energy storage device group to achieve coordinated control of the group; multiple steam energy storage devices are used to store and release steam energy; a sensor network includes a temperature sensor, a pressure sensor and a flow sensor, which are used to obtain real-time data related to the steam energy storage device group; a control center is used to train the control model, generate the control strategy and apply it to the steam energy storage device group.

[0038] S2. The thermal load prediction model combines the energy storage and release data of the steam energy storage device group to predict the load demand.

[0039] The present invention is further configured as follows: the heat load prediction model predicts the load demand based on the BP neural network prediction algorithm, and at the same time uses a genetic algorithm to optimize the BP neural network prediction algorithm.

[0040] The present invention is further configured as follows: the BP neural network prediction algorithm learns and trains the input sample data, mines the rules hidden in the data, and uses the learned rules to predict future load data. The genetic algorithm uses a parallel random search optimization method to select individuals according to the fitness function to retain individuals with strong fitness and eliminate individuals with poor fitness, while the newly formed group inherits the excellent information of the previous generation. The genetic algorithm is a global optimization algorithm used to make up for the disadvantage of the BP neural network algorithm falling into a local minimum, screen out the optimal BP weights and thresholds, and improve the prediction accuracy and convergence speed.

[0041] The present invention is further configured as follows: the BP neural network performs load demand prediction including the following sub-steps:

[0042] S21, selecting the temperature, pressure, flow data, power generation and heat load influencing factors of the steam energy storage equipment group for the previous three days as input variables;

[0043] S22, preprocessing the data, taking the predicted daily heat load as an output variable, wherein the preprocessing includes denoising, filtering and normalizing the data to improve the accuracy and availability of the data;

[0044] S23, determine the structure of the BP neural network, including the number of nodes in the input layer, hidden layer and output layer and their connection mode, use a small random value to initialize and set the parameters of the genetic algorithm, the hidden layer is one layer, and the number of nodes in the hidden layer is , where π is the number of input variables;

[0045] S24, calculating the fitness by the training error of the BP neural network, and performing crossover and mutation operations when conditions are met;

[0046] S25. When the conditions are not met, the optimal weighted network value is obtained, the error is measured, the weighted threshold is updated, and the load demand is predicted when the conditions are met.

[0047] S3. Generate a control strategy through the heat load prediction model and carry out distribution network control and optimization strategy to achieve the best operating state of the steam energy storage equipment.

[0048] The present invention is further configured as follows: the control strategy includes adjusting the power output of each device in the group to achieve the optimal operating state of the steam energy storage device, and at the same time predicting the load demand through BP neural network prediction. The control strategy can predict the optimal power output based on historical data and real-time data to meet the needs of the power grid and maximize energy utilization efficiency.

[0049] S4. Control strategy applies the control method to the steam energy storage device group to achieve coordinated control of the steam energy storage device group.

[0050] The present invention is further configured as follows: the control execution can achieve control of the equipment by adjusting the operating parameters of the equipment, such as valve opening or steam release rate, etc., and the control strategy can be adjusted in real time by monitoring the status and feedback information of each device in the group to ensure stable operation and load balance of the group, and ultimately realize the conversion of unstable energy of renewable energy electricity into stable steam input.

[0051] In summary, the present invention transmits unstable, unstable and non-continuous renewable energy power to a group of steam energy storage devices. The thermal load prediction model uses a BP neural network prediction algorithm and is optimized by a genetic algorithm to predict the load demand. At this time, the control strategy can predict the optimal power output based on historical data and real-time data to meet the needs of the power grid and maximize energy utilization efficiency. The control method is applied to the steam energy storage device group to achieve coordinated control of the steam energy storage device group, improve the accuracy and adaptability of the control strategy, and at the same time, according to the load demand prediction, the steam demand of the end user can be compared to predict and calculate the target load steam demand. When the difference in user steam demand cannot be met, the use of renewable energy power during the day cannot meet the sufficient steam demand. During the off-peak electricity period at night, low electricity prices can be used to prepare supplementary steam and store steam to fully reduce the steam cost and make up for the shortage of renewable energy power steam, saving the electricity cost of industrial steam, and finally realizing the conversion of unstable energy of renewable energy power into stable steam input, so as to achieve stable transmission of industrial steam.

[0052] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for controlling a group of steam energy storage devices, characterized in that: The following steps are involved: S1. Transmit unstable, unstable and discontinuous renewable energy electricity to a group of steam energy storage devices; S2. The thermal load prediction model combines the energy storage and release data of the steam energy storage device group to predict the load demand; S3, generating a control strategy through a heat load prediction model while performing distribution network control and optimization strategies to achieve the best operating state of the steam energy storage equipment; S4, control strategy: applying the control method to the steam energy storage device group to achieve coordinated control of the steam energy storage device group; In step S2, the heat load prediction model predicts the load demand based on the BP neural network prediction algorithm, and uses a genetic algorithm to optimize the BP neural network prediction algorithm; In step S2, the BP neural network performs load demand prediction including the following sub-steps: S21, selecting the temperature, pressure, flow data, power generation and heat load influencing factors of the steam energy storage equipment group for the previous three days as input variables; S22, preprocessing the data and taking the predicted daily heat load as the output variable; S23, determining the structure of the BP neural network, including the number of nodes in the input layer, the hidden layer, and the output layer and their connection mode, and using a smaller random value for initialization to set the parameters of the genetic algorithm; S24, calculating the fitness by the training error of the BP neural network, and performing crossover and mutation operations when conditions are met; S25, when the conditions are not met, the optimal weight network value is obtained, and the error is measured at the same time, and the weight threshold is updated to predict the load demand when the conditions are met; In step S23, the hidden layer is one layer, and the number of hidden layer nodes is , where π is the number of input variables; In step S3, the control strategy includes adjusting the power output of each device in the group to achieve the optimal operating state of the steam energy storage device, and predicting the load demand through BP neural network prediction. The control strategy can predict the optimal power output based on historical data and real-time data to meet the needs of the power grid and maximize energy utilization efficiency.

2. A method for controlling a group of steam energy storage devices according to claim 1, characterized in that: In step S1, the unstable, unstable and non-continuous renewable new energy electricity includes solar photovoltaic power generation, wind power generation and power abandonment during the off-peak period of the power grid. The power abandonment during the off-peak period of the power grid refers to the power abandoned when the power grid operator shuts down or reduces the production of some power generation equipment in order to maintain the balance and stability of the power grid under the condition of oversupply of electricity, resulting in part of the electricity being unable to be used. The power abandonment during the off-peak period of the power grid refers to the power abandoned when the hydropower station generates electricity and cannot convert all the water energy into electricity due to the demand for electricity.

3. A method for controlling a group of steam energy storage devices according to claim 1, characterized in that: In step S1, the steam energy storage device group includes a control system and a control device, wherein the control system includes a sensor module, a model training module, a control strategy generation module and a control execution module, and the control device includes multiple steam energy storage devices, a sensor network and a control center.

4. A method for controlling a group of steam energy storage devices according to claim 3, characterized in that: In step S1, the sensor module is used to obtain historical data and real-time data related to the steam energy storage device group, the model training module is used to train and optimize the thermal load prediction model, and the control model is trained based on the historical data and real-time data. The control strategy generation module is used to generate a control strategy according to the control model, including adjusting the power output of each device in the group to achieve load balance of the group. The control execution module is used to apply the control strategy to the steam energy storage device group to achieve coordinated control of the group. The multiple steam energy storage devices are used to store and release steam energy. The sensor network includes temperature sensors, pressure sensors and flow sensors, which are used to obtain real-time data related to the steam energy storage device group. The control center is used to train the control model, generate the control strategy and apply it to the steam energy storage device group.

5. A method for controlling a group of steam energy storage devices according to claim 1, characterized in that: In step S2, the BP neural network prediction algorithm learns and trains the input sample data, mines the rules hidden in the data, and uses the learned rules to predict future load data. The genetic algorithm uses a parallel random search optimization method to select individuals according to the fitness function to retain individuals with strong fitness and eliminate individuals with poor fitness. The newly formed group inherits the excellent information of the previous generation. The genetic algorithm is a global optimization algorithm used to make up for the shortcomings of the BP neural network algorithm falling into local minima and screen out the optimal BP weights and thresholds.

6. A method for controlling a group of steam energy storage devices according to claim 1, characterized in that: In step S22, the preprocessing includes denoising, filtering and normalizing the data to improve the accuracy and usability of the data.

Images