A Method for Flexibly Allocating and Supplying Heat with Cascade Thermal Energy in a Cogeneration Unit

Abstract

The present invention discloses a method for flexibly allocating and supplying heat in a cascade heat energy of a cogeneration unit, which includes the following steps: constructing a long short-term memory artificial neural network model; obtaining the heat supply data of the previous moment and inputting it into the trained long short-term memory artificial neural network model to obtain the predicted heat supply flow value of the high-pressure heater drain at the current moment; using the high-pressure heater drain with the predicted flow value and the extraction steam of the cogeneration unit for temperature difference mixing to obtain the heat energy corresponding to the target temperature of industrial users; using the heat energy corresponding to the target temperature of industrial users obtained by the temperature difference mixing to supply heat to users. The present invention can not only reduce the loss of available energy by using the small temperature difference mixing of the high-pressure heater drain and the extraction steam in the regenerative system, but also output the predicted heat supply flow value of the high-pressure heater drain by using the constructed and trained long short-term memory artificial neural network model, which can further reduce the loss of available energy and better meet the usage requirements of enterprises.

Description

Technical Field

[0001] The present invention relates to the technical field of heat supply for thermal power units, and more specifically, to a method for flexibly allocating cascade heat energy in a cogeneration unit for heat supply. Background Art

[0002] With the rise of fuel costs and the decline of power generation utilization hours, the power generation profit is getting lower and lower, while heat supply has become a new profit growth point. Against this background, more and more pure condensing units have been transformed into cogeneration units, that is, from the original single power generation mode to the power generation + heat supply mode. Cogeneration is an advanced concept based on the cascade utilization of heat energy, which is an energy-efficient production method that takes into account both electric energy and heat energy, and has obvious advantages in terms of economy and environment. Compared with the separate production of heat and power, the high-grade heat energy in cogeneration is used for power generation, and the low-grade heat energy is used for heat supply - the chemical energy of fuel is converted into high-grade heat energy, and the working medium carrying heat energy enters the steam turbine to do work and generate electricity, and the low-grade heat energy after doing work is used for heat supply.

[0003] At present, the current capacity grades of cogeneration units include 135 MW and below, 200 MW, 300 MW, 600 MW, and 1000 MW, etc., and the heat supply modes include back-pressure machine heat supply, high back-pressure heat supply, zero output of low-pressure cylinder heat supply, and extraction steam heat supply, etc. During the operation of the cogeneration unit, since the extraction steam temperature may exceed the required temperature of industrial users, it is necessary to carry out spray cooling. However, the desuperheating water of the traditional desuperheater is generally taken from the outlet of the condensate pump, and the temperature is close to normal temperature. Therefore, it is easy to cause the loss of the ability of heat flow to do work due to the large temperature difference mixing of high-temperature extraction steam and condensate water of the unit close to normal temperature, that is loss or exergy loss. Therefore, in the present invention, the No. 3 high-pressure heater drain in the regenerative system is selected to be mixed with the extraction steam at a small temperature difference to reduce the exergy loss, and a method for flexibly allocating cascade heat energy in a cogeneration unit for heat supply is proposed. Summary of the Invention

[0004] (1) Technical Problems to be Solved

[0005] In view of the deficiencies of the prior art, the present invention provides a method for flexibly allocating cascade heat energy in a cogeneration unit for heat supply, which has the advantage of accurately using the high-pressure heater drain with predicted flow rate to be mixed with the extraction steam at a temperature difference to further reduce the exergy loss, and thus solves the problems in the background art.

[0006] (2) Technical Solutions

[0007] To achieve the above-mentioned advantage of accurately using the high-pressure heater drain with predicted flow rate to be mixed with the extraction steam at a temperature difference to further reduce the exergy loss, the specific technical solutions adopted by the present invention are as follows:

[0008] A method for flexibly allocating cascade heat energy of a combined heat and power unit, applicable to the high-pressure heater drain heat supply allocation system, the method comprising the following steps:

[0009] S1. Construct a long short-term memory artificial neural network model based on the heat supply data of the high-pressure heater drain heat supply allocation system operating in the historical database and train it;

[0010] S2. Obtain the heat supply data of the previous moment and input it into the trained long short-term memory artificial neural network model to obtain the predicted value of the heat supply flow of the high-pressure heater drain at the current moment;

[0011] S3. Use the high-pressure heater drain with the predicted flow at the current moment and the extraction steam of the combined heat and power unit for temperature difference mixing to obtain the heat energy corresponding to the target temperature of industrial users;

[0012] S4. Use the heat energy corresponding to the target temperature of industrial users obtained by temperature difference mixing to supply heat to users, realizing flexible heat supply allocation for target users.

[0013] Further, the high-pressure heater drain heat supply allocation system includes a medium-pressure steam turbine, a deaerator, a high-pressure heater, a primary desuperheater and a secondary desuperheater;

[0014] The medium-pressure steam turbine is connected to the deaerator and the high-pressure heater respectively through the extraction steam heat supply pipeline, and two groups of desuperheating water pipelines are respectively arranged on the extraction steam heat supply pipeline output to the off-site industrial extraction steam network, and a primary desuperheater and a secondary desuperheater are respectively arranged on the two groups of desuperheating water pipelines;

[0015] The deaerator and the high-pressure heater are connected through the high-pressure heater drain pipeline, and the desuperheating water pipelines are all connected to the high-pressure heater drain pipeline through the connecting pipeline;

[0016] A number of control valves are arranged on the extraction steam heat supply pipeline, the desuperheating water pipeline, the high-pressure heater drain pipeline and the connecting pipeline, and the extraction steam heat supply pipeline and the desuperheating water pipeline are connected through a vacuum pump.

[0017] Further, the extraction steam heat supply pipeline, the high-pressure heater drain pipeline and the connecting pipeline are all heat-insulated pipelines, the desuperheating water pipeline is a non-heat-insulated pipeline, and the control valves include check valves, globe valves and flow regulating valves.

[0018] Further, temperature sensors, pressure gauges and flow meters are arranged on the extraction steam heat supply pipeline, the desuperheating water pipeline and the connecting pipeline.

[0019] Further, the heat supply data includes heat supply parameters of the heat supply station and heat supply parameters of users; among them, the heat supply parameters of the heat supply station include the temperature, flow rate of the extraction steam of the combined heat and power unit and the temperature, flow rate of the high-pressure heater drain, and the heat supply parameters of users include the target temperature of industrial users.

[0020] Further, constructing and training a long short-term memory artificial neural network model based on the heating data of the high-pressure heater drain heating distribution system in the historical database includes the following steps:

[0021] S11. Construct a long short-term memory artificial neural network model;

[0022] S12. Obtain the heating data of the high-pressure heater drain heating distribution system in the historical database, and preprocess the collected heating data;

[0023] S13. Use the preprocessed heating data as the input sequence of the long short-term memory artificial neural network model;

[0024] S14. Train the long short-term memory artificial neural network model based on the input sequence data to obtain the trained long short-term memory artificial neural network model.

[0025] Further, the preprocessing includes outlier removal, filling in missing values, data classification, and data sorting.

[0026] Further, training the long short-term memory artificial neural network model based on the input sequence data also includes the following steps:

[0027] For the case of insufficient training, increase the number of nodes in the network or increase the training cycle of the network to achieve the training effect;

[0028] For the case of overfitting, reduce or control the training cycle, and stop training the network before the data shows an inflection point to achieve the training effect.

[0029] Further, obtaining the heating data of the previous moment and inputting it into the trained long short-term memory artificial neural network model to obtain the predicted value of the heating flow of the high-pressure heater drain at the current moment includes the following steps:

[0030] S21. Obtain the heating data of the previous moment, and determine whether the temperature of the extraction steam of the combined heat and power unit in the heating data exceeds the requirements of industrial users. If so, execute S22; if not, use the extraction steam at this temperature to heat industrial users;

[0031] S22. Use the temperature, flow rate of the extraction steam of the combined heat and power unit, the temperature of the high-pressure heater drain, and the target temperature of industrial users in the heating data as the input parameters of the long short-term memory artificial neural network model;

[0032] S23. Use the trained long short-term memory artificial neural network model to output the predicted value of the heating flow of the high-pressure heater drain at the current moment.

[0033] Further, the steps of obtaining the thermal energy corresponding to the target temperature of industrial users by performing temperature difference mixing on the high-pressure heater drain water with the predicted flow value at the current moment and the extraction steam of the cogeneration unit are as follows:

[0034] S31. Connect the high-pressure heater drain water with the predicted flow into the primary desuperheater, keep the condensate pipe of the secondary desuperheater closed, and perform temperature difference mixing on the high-pressure heater drain water in the primary desuperheater and the extraction steam of the cogeneration unit to obtain the thermal energy corresponding to the target temperature of industrial users;

[0035] S32. Determine whether the regulation of industrial gas supply parameters by the primary desuperheater exceeds the standard. If not, continue to execute S31. If so, promptly start the secondary desuperheater to match the condensate for temperature reduction, and use the temperature sensor to monitor the temperature of the output thermal energy in real time until the temperature of the output thermal energy reaches the target temperature of industrial users.

[0036] (III) Beneficial Effects

[0037] Compared with the prior art, the present invention provides a method for flexible allocation of cascade thermal energy of a cogeneration unit for heating, which has the following beneficial effects:

[0038] (1) The present invention can not only reduce the loss of available energy by performing small temperature difference mixing on the high-pressure heater drain water and extraction steam in the regenerative system, but also use the constructed and trained long short-term memory artificial neural network model to output the predicted flow value of the high-pressure heater drain water, so that the high-pressure heater drain water with the predicted flow can be accurately used to perform temperature difference mixing with the extraction steam to obtain the thermal energy corresponding to the target temperature of industrial users, thereby further reducing the loss of available energy and better meeting the usage requirements of enterprises.

[0039] (2) By replacing the traditional mixing and cooling method of industrial extraction steam matching condensate with the mixing and cooling method of industrial extraction steam matching high-pressure heater drain water, the present invention can effectively improve the overall available energy efficiency, can be widely promoted in cogeneration units, and has broad application prospects. Description of the Drawings

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0041] Figure 1 is a flowchart of the method for flexible allocation of cascade thermal energy of a cogeneration unit according to an embodiment of the present invention;

[0042] Figure 2It is a schematic structural diagram of a high-pressure heater drain heating distribution system in the cascade heat energy flexible distribution heating method for a combined heat and power unit according to an embodiment of the present invention.

[0043] In the figure:

[0044] 1. Intermediate-pressure steam turbine; 2. Deaerator; 3. High-pressure heater; 4. Primary desuperheater; 5. Secondary desuperheater; 6. Extraction steam heating pipeline; 7. Desuperheating water pipeline; 8. High-pressure heater drain pipeline; 9. Connecting pipeline. Specific embodiments

[0045] To further illustrate the embodiments, the present invention provides accompanying drawings, which are part of the disclosure of the present invention. They are mainly used to illustrate the embodiments and can be combined with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, those of ordinary skill in the art should be able to understand other possible implementation manners and the advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are usually used to represent similar components.

[0046] According to an embodiment of the present invention, a method for flexible distribution of cascade heat energy in a combined heat and power unit is provided.

[0047] Now, the present invention will be further described in combination with the accompanying drawings and specific embodiments. As Figure 1 - Figure 2 shown, the method for flexible distribution of cascade heat energy in a combined heat and power unit according to an embodiment of the present invention is applicable to a high-pressure heater drain heating distribution system. The high-pressure heater drain heating distribution system includes an intermediate-pressure steam turbine 1, a deaerator 2, a high-pressure heater 3, a primary desuperheater 4, and a secondary desuperheater 5;

[0048] The intermediate-pressure steam turbine 1 is connected to the deaerator 2 and the high-pressure heater 3 respectively through an extraction steam heating pipeline 6, and two groups of desuperheating water pipelines 7 are respectively arranged on the extraction steam heating pipeline 6 leading to the off-site industrial extraction steam network. A primary desuperheater 4 and a secondary desuperheater 5 are respectively arranged on the two groups of desuperheating water pipelines 7; The deaerator 2 and the high-pressure heater 3 are connected through a high-pressure heater drain pipeline 8, and the desuperheating water pipelines 7 are all connected to the high-pressure heater drain pipeline 8 through a connecting pipeline 9; A number of control valves are arranged on the extraction steam heating pipeline 6, the desuperheating water pipeline 7, the high-pressure heater drain pipeline 8, and the connecting pipeline 9, and the extraction steam heating pipeline 6 and the desuperheating water pipeline 7 are connected through a vacuum pump.

[0049] The extraction steam heating pipeline 6, the high-pressure heater drain pipeline 8, and the connecting pipeline 9 are all heat-insulated pipelines, and the desuperheating water pipeline 7 is a non-heat-insulated pipeline. The control valves include check valves, globe valves, and flow regulating valves; Temperature sensors, pressure gauges, and flow meters are arranged on the extraction steam heating pipeline 6, the desuperheating water pipeline 7, and the connecting pipeline 9.

[0050] The method for flexible distribution of cascade heat energy in the combined heat and power unit includes the following steps:

[0051] S1. Construct a long short - term memory artificial neural network model based on the heating data of the high - pressure heater drain heating distribution system in the historical database and train it;

[0052] Among them, constructing a long short - term memory artificial neural network model based on the heating data of the high - pressure heater drain heating distribution system in the historical database and training it includes the following steps:

[0053] S11. Construct a long short - term memory artificial neural network model;

[0054] S12. Obtain the heating data of the high - pressure heater drain heating distribution system in the historical database and pre - process the collected heating data. Specifically, the heating data includes the heating parameters of the heat sub - station and the heating parameters of the users. Among them, the heating parameters of the heat sub - station include the temperature and flow rate of the extraction steam of the cogeneration unit and the temperature and flow rate of the high - pressure heater drain, etc., and the heating parameters of the users include the target temperature of industrial users, etc.; the pre - processing includes outlier removal, filling in missing values, data classification, and data sorting;

[0055] S13. Use the pre - processed heating data as the input sequence of the long short - term memory artificial neural network model;

[0056] S14. Train the long short - term memory artificial neural network model based on the input sequence data to obtain the trained long short - term memory artificial neural network model.

[0057] Specifically, training the long short - term memory artificial neural network model based on the input sequence data also includes the following steps:

[0058] For the case of insufficient training, achieve the training effect by increasing the nodes in the network or increasing the training cycle of the network;

[0059] For the case of overfitting, achieve the training effect by reducing or controlling the training cycle and stopping the training of the network before the data shows an inflection point.

[0060] S2. Obtain the heating data of the previous moment and input it into the trained long short - term memory artificial neural network model to obtain the predicted value of the heating flow rate of the high - pressure heater drain at the current moment;

[0061] Among them, obtaining the heating data of the previous moment and inputting it into the trained long short - term memory artificial neural network model to obtain the predicted value of the heating flow rate of the high - pressure heater drain at the current moment includes the following steps:

[0062] S21. Obtain the heating data of the previous moment and determine whether the temperature of the extraction steam of the cogeneration unit in the heating data exceeds the requirements of industrial users. If so, execute S22; if not, use the extraction steam at this temperature to supply heat to industrial users;

[0063] S22. Use the temperature, flow rate of the extraction steam of the cogeneration unit, the temperature of the high-pressure heater drain, and the target temperature of the industrial users in the heating data as the input parameters of the long short-term memory artificial neural network model;

[0064] S23. Use the trained long short-term memory artificial neural network model to output the predicted value of the heating flow rate of the high-pressure heater drain at the current moment.

[0065] In this embodiment, the high-pressure heater drain is selected as the water source for adjusting the industrial extraction steam. First, the temperature of the high-pressure heater drain is nearly 180°C, which is relatively high and meets the water temperature requirements. Second, the water quality of the high-pressure heater drain is relatively stable, and no secondary problems will be caused by nozzle blockage of the desuperheater due to water quality problems during long-term operation. Third, the pressure of the high-pressure heater drain is 1.5 MPa, which is close to the outlet pressure of the condensate pump and meets the operation requirements of the desuperheater. Fourth, the industrial steam pipeline is close to the deaerator. Connecting a branch pipe from the high-pressure heater drain pipeline to the industrial steam pipeline requires less material and lower construction cost.

[0066] S3. Mix the high-pressure heater drain with the extraction steam of the cogeneration unit at the predicted flow rate at the current moment by temperature difference to obtain the heat energy corresponding to the target temperature of the industrial users;

[0067] Among them, mixing the high-pressure heater drain with the extraction steam of the cogeneration unit at the predicted heating flow rate at the current moment by temperature difference to obtain the heat energy corresponding to the target temperature of the industrial users includes the following steps:

[0068] S31. Connect the high-pressure heater drain with the predicted flow rate to the primary desuperheater, and keep the condensate water pipeline of the secondary desuperheater closed. Use the high-pressure heater drain in the primary desuperheater to mix with the extraction steam of the cogeneration unit by temperature difference to obtain the heat energy corresponding to the target temperature of the industrial users;

[0069] S32. Judge whether the industrial gas supply parameters regulated by the primary desuperheater exceed the standard. If not, continue to execute S31. If so, immediately start the secondary desuperheater to match the condensate water for temperature reduction, and use the temperature sensor to monitor the temperature of the output heat energy in real time until the temperature of the output heat energy reaches the target temperature of the industrial users.

[0070] In this embodiment, since the new research working condition does not change the secondary desuperheater, the desuperheating water pipeline, the operation mode and the control logic, the secondary desuperheater still maintains the original matching factor and the original regulation ability for regulating the industrial steam index after the new adjustment and is not affected by any new adjustment.

[0071] In addition, when the high-pressure heater drain fails to regulate the industrial steam index and the secondary desuperheating water is ineffective, the primary desuperheater can close the high-pressure heater drain valve and open the original condensate water pipeline for water supply. That is, the primary desuperheater can flexibly enable the new working condition or return to the original operation condition at any time.

[0072] S4. Use the thermal energy corresponding to the target temperature of industrial users obtained by temperature difference mixing to supply heat to users, so as to achieve flexible allocation of heat supply for target users.

[0073] In summary, with the above technical solutions of the present invention, the present invention can not only use the high-pressure heater drain water and extraction steam in the regenerative system for small temperature difference mixing to reduce the loss of available energy, but also use the constructed and trained long short-term memory artificial neural network model to output the flow prediction value of the high-pressure heater drain water, so that the high-pressure heater drain water with the predicted flow can be accurately used for temperature difference mixing with the extraction steam to obtain the thermal energy corresponding to the target temperature of industrial users, thereby further reducing the loss of available energy and better meeting the usage requirements of enterprises.

[0074] In addition, by replacing the traditional mixing and cooling method of industrial extraction steam matching condensate with the mixing and cooling method of industrial extraction steam matching high-pressure heater drain water, the present invention can effectively improve the overall available energy efficiency, can be widely promoted in cogeneration units, and has broad application prospects.

[0075] In the present invention, unless otherwise clearly specified and defined, terms such as "installation", "setting", "connection", "fixation", "swivel connection" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or integrated; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the internal communication of two components or the interaction relationship between two components. Unless otherwise clearly defined, for those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0076] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for flexibly allocating and supplying heat with cascaded thermal energy of a combined heat and power unit, characterized in that, The cascade heat energy flexible dispatching and heating method for the cogeneration unit is applicable to the high-pressure heater drain heating dispatching system, and this method includes the following steps: S1. Construct and train a long short-term memory artificial neural network model based on the heating data of the high-pressure heater drain heating dispatching system operation in the historical database; The constructing and training of the long short-term memory artificial neural network model based on the heating data of the high-pressure heater drain heating dispatching system operation in the historical database includes the following steps: S11. Construct a long short-term memory artificial neural network model; S12. Obtain the heating data of the high-pressure heater drain heating dispatching system operation in the historical database, and preprocess the collected heating data; S13. Use the preprocessed heating data as the input sequence of the long short-term memory artificial neural network model; S14. Train the long short-term memory artificial neural network model based on the input sequence data to obtain the trained long short-term memory artificial neural network model; S2. Obtain the heating data of the previous moment and input it into the trained long short-term memory artificial neural network model to obtain the flow prediction value of the high-pressure heater drain at the current moment; The obtaining of the heating data of the previous moment and inputting it into the trained long short-term memory artificial neural network model to obtain the flow prediction value of the high-pressure heater drain at the current moment includes the following steps: S21. Obtain the heating data of the previous moment, and judge whether the temperature of the extraction steam of the cogeneration unit in the heating data exceeds the requirements of industrial users. If so, execute S22. If not, use the extraction steam at this temperature to heat industrial users; S22. Use the temperature, flow rate of the extraction steam of the cogeneration unit, the temperature of the high-pressure heater drain, and the target temperature of industrial users in the heating data as the input parameters of the long short-term memory artificial neural network model; S23. Use the trained long short-term memory artificial neural network model to output the flow prediction value of the high-pressure heater drain at the current moment; S3. Use the high-pressure heater drain corresponding to the flow prediction value of the high-pressure heater drain at the current moment and the extraction steam of the cogeneration unit for temperature difference mixing to obtain the heat energy corresponding to the target temperature of industrial users; The using of the high-pressure heater drain corresponding to the flow prediction value of the high-pressure heater drain at the current moment and the extraction steam of the cogeneration unit for temperature difference mixing to obtain the heat energy corresponding to the target temperature of industrial users includes the following steps: S31. Connect the high-pressure heater drain with the predicted flow rate to the primary desuperheater, and keep the condensate pipeline of the secondary desuperheater closed. Use the high-pressure heater drain in the primary desuperheater and the extraction steam of the cogeneration unit for temperature difference mixing to obtain the heat energy corresponding to the target temperature of industrial users; S32. Judge whether the regulated industrial gas supply parameters of the primary desuperheater exceed the standard. If not, continue to execute S31. If so, promptly start the secondary desuperheater to match the condensate for temperature reduction, and use the temperature sensor to continuously monitor the temperature of the output heat energy until the temperature of the output heat energy reaches the target temperature of industrial users; S4. Use the heat energy corresponding to the target temperature of industrial users obtained by temperature difference mixing to heat users, realizing flexible dispatching and heating for target users.

2. A method for flexibly allocating heat energy in a cascade manner for a combined heat and power unit according to claim 1, characterized in that, The high-pressure heater drain heating distribution system includes a medium-pressure steam turbine (1), a deaerator (2), a high-pressure heater (3), a primary desuperheater (4), and a secondary desuperheater (5); The medium-pressure steam turbine (1) is connected to the deaerator (2) and the high-pressure heater (3) respectively through a steam extraction heating pipeline (6), and two groups of desuperheating water pipelines (7) are respectively arranged on the steam extraction heating pipeline (6) output to the off-site industrial steam extraction network. The primary desuperheater (4) and the secondary desuperheater (5) are respectively arranged on the two groups of desuperheating water pipelines (7); The deaerator (2) and the high-pressure heater (3) are connected through a high-pressure heater drain pipeline (8), and the desuperheating water pipelines (7) are all connected to the high-pressure heater drain pipeline (8) through connecting pipelines (9); A number of control valves are arranged on the steam extraction heating pipeline (6), the desuperheating water pipeline (7), the high-pressure heater drain pipeline (8), and the connecting pipeline (9), and the steam extraction heating pipeline (6) and the desuperheating water pipeline (7) are connected through a vacuum pump.

3. A method for flexibly allocating and supplying heat with cascaded thermal energy of a cogeneration unit according to claim 2, characterized in that, The steam extraction heating pipeline (6), the high-pressure heater drain pipeline (8), and the connecting pipeline (9) are all heat-insulated pipelines, the desuperheating water pipeline (7) is a non-heat-insulated pipeline, and the control valves include check valves, globe valves, and flow regulating valves.

4. A method for flexibly allocating and supplying heat with cascade heat energy of a combined heat and power unit according to claim 2, characterized in that, Temperature sensors, pressure gauges, and flow meters are arranged on the steam extraction heating pipeline (6), the desuperheating water pipeline (7), and the connecting pipeline (9).

5. A method for flexibly allocating and supplying heat in a cascade heat energy of a cogeneration unit according to claim 1, characterized in that, The heating data includes heat supply parameters of the heat supply station and heat supply parameters of users; among them, the heat supply parameters of the heat supply station include the temperature, flow rate of the steam extraction of the cogeneration unit, and the temperature, flow rate of the high-pressure heater drain, and the heat supply parameters of users include the target temperature of industrial users.

6. A method for flexibly allocating and supplying heat with cascaded heat energy of a cogeneration unit according to claim 1, characterized in that, The preprocessing includes outlier removal, filling of missing points, data classification, and data sorting.

7. A method for flexibly allocating and supplying heat from cascade heat energy of a combined heat and power unit according to claim 1, characterized in that The training of the long short-term memory artificial neural network model based on the input sequence data further includes the following steps: For the case of insufficient training, the training effect is achieved by increasing the nodes in the network or increasing the training cycle of the network; For the case of overfitting, the training effect is achieved by reducing or controlling the training cycle and stopping the training of the network before the data shows an inflection point.

Images