A heat control method based on cogeneration

Abstract

The present invention discloses a heat control method based on cogeneration, which classifies environmental data to obtain classified data; obtains predicted demand; adjusts the heat supplied by cogeneration according to the predicted demand; and controls the equipment involved in the method according to the actual demand and the heat supplied by cogeneration. It relates to the field of cogeneration control technology, and solves the technical problem that the heat supply and the user-side heat load demand are inconsistent, which leads to repeated heat surplus and heat gap in a short period of time, thereby generating heat waste; by predicting future demand, the heat supply of thermal power output is adjusted according to the predicted demand, even if there is a deviation between the predicted demand and the actual demand, resulting in a difference between the actual demand and the heat supplied by cogeneration, the heat energy can be absorbed or supplemented through geothermal wells at this time, thereby reducing energy consumption and significantly improving the utilization efficiency of geothermal wells.

Description

Technical Field

[0001] The invention belongs to the technical field of cogeneration control, and in particular is a heat control method based on cogeneration. Background Art

[0002] Cogeneration is a major way of urban central heating and a technical means to improve energy utilization. It uses heat engines or power plants to generate electricity and useful heat at the same time. This technology can not only significantly improve fuel utilization, but also is an important measure to save energy and enhance the functions of urban infrastructure. It is also an important means to ensure heating safety and promote green and low-carbon development.

[0003] However, there are still some problems in heat regulation. When the heat supply of cogeneration can be set according to demand, the load demand will change, so the changes in the demand for heating and power generation are not synchronized, resulting in inconsistency between the output heat and the thermal load demand at the user end, resulting in repeated heat surpluses and heat gaps in the short term, resulting in heat waste and failure to maximize economic benefits. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a heat control method based on cogeneration of heat and power, which is used to solve the technical problem that the heating heat is inconsistent with the heat load demand at the user end, thereby resulting in repeated heat surplus and heat gap in a short period of time, thereby generating heat waste. The present invention predicts future demand and adjusts the supply heat of thermal power output according to the predicted demand, thereby reducing the phenomenon of a large amount of heat energy being generated but not utilized to a certain extent. Even if there is a deviation between the predicted demand and the actual demand, the heat energy can be absorbed or supplemented through geothermal wells to solve the above problem.

[0005] To achieve the above object, a first aspect of the present invention provides a heat control method based on cogeneration, comprising the following steps:

[0006] Step 1: Obtain time data and environmental data, and classify the environmental data to obtain classified data; wherein the classified data includes abnormal data and normal data;

[0007] Step 2: Obtain forecast demand based on time data, classification data and environmental data; adjust the heat supply of cogeneration according to the forecast demand;

[0008] Step 3: Determine whether the predicted demand is greater than the heat supplied by the cogeneration; if not, control the peak-shaving boiler and heat pump unit to stop and store the excess heat supply in the geothermal well; if yes, control the peak-shaving boiler and heat pump unit to operate and connect the heat in the geothermal well to the peak-shaving boiler.

[0009] First, the demand in the cogeneration area is predicted, and then the cogeneration supply is adjusted according to the predicted demand. If there is a deviation between the predicted demand and the supplied heat, energy can be stored or released through geothermal wells to meet the demand. This method can greatly avoid waste and insufficient supply problems and maximize economic benefits.

[0010] It should be noted that the time data may be a certain time period within a set period.

[0011] Preferably, the step of classifying the environmental data to obtain classified data includes:

[0012] Extract environmental data, and compare the sub-data in the environmental data with the corresponding range thresholds respectively; if the sub-data in the environmental data are all within the corresponding range thresholds, the environmental data are marked as normal data; otherwise, the environmental data are marked as abnormal data; wherein the environmental data includes ambient temperature, ambient humidity and wind speed, the range threshold corresponding to the ambient temperature is the temperature range, the range threshold corresponding to the ambient humidity is the humidity range, the range threshold corresponding to the wind speed is the wind speed range, and the range threshold corresponding to the sub-data in the environmental data is obtained through historical data in the database.

[0013] By dividing the environmental data into abnormal data and normal data, one of the input data can be improved in the subsequent prediction process, which can improve the subsequent prediction accuracy.

[0014] Preferably, the range threshold corresponding to the sub-data in the environmental data is obtained through historical data in a database, including:

[0015] S1: extracting a number of historical loads and environmental data corresponding to the historical loads from a database;

[0016] S2: Sort several historical loads from large to small to obtain load ranking, mark the loads corresponding to the quantiles n% and 1-n% in the load ranking as high quantiles and low quantiles respectively, and mark the loads between the high quantiles and the low quantiles as the load range; where n is a positive number;

[0017] S3: Extract environmental data corresponding to the historical load within the load range, mark the range formed by the maximum temperature and the minimum temperature in the environmental data as the temperature range, mark the range formed by the maximum humidity and the minimum humidity as the humidity range, and mark the range formed by the maximum wind speed and the minimum wind speed as the wind speed range.

[0018] It should be noted that the specific value of n can be determined by the total amount of data of the historical load. When there is no corresponding load for quantile n% in the load sorting, the load corresponding to quantile n% can be expressed by taking the average of two adjacent loads, and the historical load and the environmental data corresponding to the historical load are obtained from the local database.

[0019] Based on the historical data in the local database, the historical data is used as the data basis for obtaining the range threshold, which can more realistically reflect the actual situation of the cogeneration supply area.

[0020] Preferably, the obtaining of predicted demand based on time data, classification data and environmental data includes:

[0021] Mark the abnormal data in the classified data with a unique characteristic number, and mark the normal data with a unique non-characteristic number;

[0022] The time data, the numbers corresponding to the classification data and the environmental data are input into the demand forecasting model to obtain the predicted load, wherein the demand forecasting model is obtained based on the training of the artificial intelligence model.

[0023] Preferably, the demand forecasting model is obtained based on artificial intelligence model training, including:

[0024] Obtain historical load, time data corresponding to the historical load, and environmental data from a database; classify the environmental data based on the historical load to obtain classified training data; wherein the classified training data includes normal training data and abnormal training data; mark the abnormal training data with a unique characteristic number, and mark the normal training data with a unique non-characteristic number;

[0025] The historical load, the time data, environmental data and test data corresponding to the historical load are integrated into several training data groups and test data groups; the artificial intelligence model is trained by the several training data groups, and then the trained artificial intelligence model is tested by the test data group, and the artificial intelligence model is modified according to the test results to obtain a demand forecasting model whose input is the numbers corresponding to the time data, environmental data and classification data, and whose output is the predicted demand.

[0026] Preferably, the step of adjusting the heat supply of cogeneration according to the predicted demand comprises:

[0027] The current CHP supply is counted; the predicted demand is compared with the current CHP supply; when the predicted demand is less than the current CHP supply, the operating parameters of the CHP equipment are adjusted to reduce the heat supply of the CHP equipment; when the predicted demand is greater than the current CHP supply, the operating parameters of the CHP equipment are adjusted to increase the heat supply of the CHP equipment; when the predicted demand is equal to the current CHP supply, the operating parameters of the CHP equipment are maintained.

[0028] Preferably, storing the excess supplied heat in a geothermal well comprises:

[0029] The plate heat exchanger is started to exchange heat with the geothermal water in the geothermal well, and the geothermal water is re-injected into the geothermal well.

[0030] Preferably, the step of connecting the heat in the geothermal well to the peak-shaving boiler includes:

[0031] The hot water in the geothermal well and the water source in the cogeneration are passed through the plate heat exchanger for heat exchange, and the geothermal water after heat exchange is sent to the evaporation side of the heat pump unit to extract heat;

[0032] After heat is extracted, the geothermal water is reinjected into the geothermal well. The water from the outlet side of the plate heat exchanger and the condensing side of the heat pump unit is mixed and passed into the peak-shaving boiler for heating. The heated water is then sent to the water supply network of the cogeneration system.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] 1. In the present invention, the current environmental data is first classified, and the future demand is predicted based on the classification results, the current time data and the environmental data. The heat supply of the thermal power output is adjusted according to the predicted demand, thereby reducing the phenomenon that a large amount of heat energy is generated but not utilized to a certain extent. Even if there is a deviation between the predicted demand and the actual demand, resulting in a difference between the predicted demand and the heat supply of the cogeneration, the heat energy can be absorbed or supplemented through the geothermal well, thereby reducing energy consumption, significantly improving the utilization efficiency of the geothermal well, and maximizing the economic benefits.

[0035] 2. In the present invention, by distinguishing environmental data to determine whether it is abnormal data, and then distinguishing abnormal data from normal data by different numbers, as one of the inputs of the demand forecasting model, the trained demand forecasting model can be more in line with the actual situation and improve the accuracy of predicted demand. Moreover, when classifying environmental data, it is based on the range threshold, and the range threshold is obtained through historical data in the database, which is more suitable for edge computing and can further improve its prediction accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0037] Figure 1 It is a schematic flow chart of the heat control method based on cogeneration of heat and power of the present invention;

[0038] Figure 2 A schematic diagram of a process for obtaining a range threshold value corresponding to sub-data in environmental data according to the present invention;

[0039] Figure 3 It is a connection diagram of the heat control equipment of the cogeneration of heat and power in the present invention. DETAILED DESCRIPTION

[0040] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] See also Figure 1-Figure 3 The first embodiment of the present invention provides a heat control method based on cogeneration, comprising the following steps:

[0042] Step 1: Obtain time data and environmental data, and classify the environmental data to obtain classified data; wherein the classified data includes abnormal data and normal data; wherein the environmental data includes environmental temperature, environmental humidity and wind speed;

[0043] Step 2: Obtain forecast demand based on time data, classification data and environmental data; adjust the heat supply of cogeneration according to the forecast demand;

[0044] Step 3: Determine whether the predicted demand is greater than the heat supplied by the cogeneration; if not, control the peak-shaving boiler and heat pump unit to stop and store the excess heat supply in the geothermal well; if yes, control the peak-shaving boiler and heat pump unit to operate and connect the heat in the geothermal well to the peak-shaving boiler.

[0045] The time data recorded above may be a certain time period within a set period. For example, the time data may be 10:00-11:00 a.m. on the second working day of the 50th week of 2023.

[0046] In this embodiment, the environmental data is classified to obtain classified data, including:

[0047] A1: Extract ambient temperature, ambient humidity and wind speed from environmental data;

[0048] A2: Compare the ambient temperature in the ambient data with the temperature range; if the ambient temperature in the ambient data is within the temperature range, mark the ambient data as normal data; otherwise, jump to A3;

[0049] A3: Compare the ambient humidity in the ambient data with the humidity range; if the ambient humidity in the ambient data is within the humidity range, mark the ambient data as normal data; otherwise, jump to A4;

[0050] A4: Compare the wind speed in the environmental data with the wind speed range; if the wind speed in the environmental data is within the wind speed range, mark the environmental data as normal data; otherwise, mark the environmental data as abnormal data.

[0051] Specifically, the temperature range, humidity range and wind speed range are obtained through historical data in the database.

[0052] By first classifying the environmental data according to specific values ​​and marking the environmental data, the data type can be used as one of the input factors when training the subsequent demand forecasting model to improve the accuracy of the forecast. The temperature range, humidity range and wind speed range can be obtained through historical data to better fit the actual use of the cogeneration.

[0053] Furthermore, the temperature range, humidity range and wind speed range are obtained through historical data in the database, including:

[0054] S1: extracting a number of historical loads and environmental data corresponding to the historical loads from a database;

[0055] S2: Sort several historical loads from large to small to obtain a load ranking, mark the loads corresponding to the quantiles of 98.5% and 1.5% in the load ranking as high quantiles and low quantiles respectively, and mark the loads between the high quantiles and the low quantiles as the load range;

[0056] S3: Extract environmental data corresponding to the historical load within the load range, mark the range formed by the maximum temperature and the minimum temperature in the environmental data as the temperature range, mark the range formed by the maximum humidity and the minimum humidity as the humidity range, and mark the range formed by the maximum wind speed and the minimum wind speed as the wind speed range.

[0057] It should be noted that the total amount of data of the historical load is 10,000, so n in this embodiment is 98.5. When the quantile n% or 1-n% does not have a corresponding load in the load sorting, for example, 9999 pieces of data, n is still 98.5 at this time, and the average of the 9849th and 9850th pieces of data is taken. The load corresponding to the quantile 98.5% can be expressed by taking the average of two adjacent loads.

[0058] In this embodiment, the predicted demand is obtained based on time data, classification data and environmental data, including:

[0059] The abnormal data in the classified data is marked with a unique characteristic number "0", and the normal data is marked with a unique non-characteristic number "1";

[0060] The time data, the numbers corresponding to the classification data and the environmental data are input into the demand forecasting model to obtain the predicted load, wherein the demand forecasting model is obtained based on the training of the artificial intelligence model.

[0061] Furthermore, the demand forecasting model is obtained based on artificial intelligence model training, including:

[0062] Obtain historical load, time data corresponding to the historical load, and environmental data from a database; classify the environmental data based on the historical load to obtain classified training data; wherein the classified training data includes normal training data and abnormal training data; mark the abnormal training data with a unique characteristic number "0", and mark the normal training data with a unique non-characteristic number "1";

[0063] The historical load, the time data corresponding to the historical load, the environmental data and the test data are integrated into several training data groups and test data groups; the artificial intelligence model is trained by means of several training data groups, and then the trained artificial intelligence model is tested by means of the test data groups. When the difference between the load in the test data group and the predicted demand output by the trained artificial intelligence model is within an acceptable threshold range, there is no need to correct the artificial intelligence model. Otherwise, the parameters such as the hidden layer in the artificial intelligence model are adjusted until the difference between the load in the test data group and the predicted demand output by the trained artificial intelligence model is within an acceptable threshold range and accounts for 95% or more of the total test data. Then, a demand forecasting model is obtained, whose input is the numbers corresponding to the time data, environmental data and classification data, and whose output is the predicted demand.

[0064] Furthermore, the artificial intelligence model may be a RBF neural network model or a BP neural network model.

[0065] In this embodiment, the heat supplied by the cogeneration is adjusted according to the predicted demand, including:

[0066] Count the current CHP supply; compare the predicted demand with the current CHP supply; when the predicted demand is less than the current CHP supply, adjust the operating parameters of the CHP equipment to reduce the heat supply of the CHP equipment; when the predicted demand is greater than the current CHP supply, adjust the operating parameters of the CHP equipment to increase the heat supply of the CHP equipment; when the predicted demand is equal to the current CHP supply, maintain the operating parameters of the CHP equipment;

[0067] For example, if the cogeneration system mainly generates electricity through fuel combustion in a gas turbine, the guide vane opening and the fuel valve opening can be adjusted to adjust the output power of the gas turbine and parameters such as the turbine exhaust flow rate and temperature.

[0068] Specifically, the excess heat supply is stored in geothermal wells, including:

[0069] The plate heat exchanger is started to exchange heat with the geothermal water in the geothermal well, and the geothermal water is re-injected into the geothermal well.

[0070] Specifically, the heat from the geothermal well is connected to the peak-shaving boiler, including:

[0071] The hot water in the geothermal well and the water source in the cogeneration are passed through the plate heat exchanger for heat exchange, and the geothermal water after heat exchange is sent to the evaporation side of the heat pump unit to extract heat;

[0072] After heat is extracted, the geothermal water is reinjected into the geothermal well. The water from the outlet side of the plate heat exchanger and the condensing side of the heat pump unit is mixed and passed into the peak-shaving boiler for heating. The heated water is then sent to the water supply network of the cogeneration system.

[0073] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0074] The equipment involved in this method mainly includes cogeneration, peak-shaving boilers, plate heat exchangers, geothermal wells and heat pump units.

[0075] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A heat control method based on cogeneration, characterized in that: The following steps are involved: Step 1: Obtain time data and environmental data, and classify the environmental data to obtain classified data; wherein the classified data includes abnormal data and normal data; Step 2: Obtain forecast demand based on time data, classification data and environmental data; adjust the heat supply of cogeneration according to the forecast demand; Step 3: Determine whether the predicted demand is greater than the heat supplied by the cogeneration; if not, control the peak-shaving boiler and heat pump unit to stop, and store the excess heat supply in the geothermal well; if yes, control the peak-shaving boiler and heat pump unit to operate, and connect the heat in the geothermal well to the peak-shaving boiler; The step of classifying the environmental data to obtain classified data includes: Extract environmental data, and compare the sub-data in the environmental data with the corresponding range thresholds respectively; if the sub-data in the environmental data are all within the corresponding range thresholds, the environmental data are marked as normal data; otherwise, the environmental data are marked as abnormal data; wherein the environmental data includes environmental temperature, environmental humidity and wind speed, the range threshold corresponding to the environmental temperature is the temperature range, the range threshold corresponding to the environmental humidity is the humidity range, the range threshold corresponding to the wind speed is the wind speed range, and the range threshold corresponding to the sub-data in the environmental data is obtained through historical data in the database; The range threshold corresponding to the sub-data in the environmental data is obtained through historical data in a database, including: S1: extracting a number of historical loads and environmental data corresponding to the historical loads from a database; S2: Sort several historical loads from large to small to obtain load ranking, mark the loads corresponding to the quantiles n% and 1-n% in the load ranking as high quantiles and low quantiles respectively, and mark the loads between the high quantiles and the low quantiles as the load range; where n is a positive number; S3: extracting environmental data corresponding to the historical load within the load range, marking the range formed by the maximum temperature and the minimum temperature in the environmental data as the temperature range, marking the range formed by the maximum humidity and the minimum humidity as the humidity range, and marking the range formed by the maximum wind speed and the minimum wind speed as the wind speed range; The obtaining of predicted demand based on time data, classification data and environmental data includes: The abnormal data in the classified data are marked with a unique characteristic number, and the normal data are marked with a unique non-characteristic number; The time data, the numbers corresponding to the classification data and the environmental data are input into the demand forecasting model to obtain the predicted load, wherein the demand forecasting model is obtained based on the training of the artificial intelligence model.

2. A heat control method based on cogeneration according to claim 1, characterized in that: The demand forecasting model is obtained based on artificial intelligence model training, including: Obtain historical load, time data corresponding to the historical load, and environmental data from a database; classify the environmental data based on the historical load to obtain classified training data; wherein the classified training data includes normal training data and abnormal training data; mark the abnormal training data with a unique characteristic number, and mark the normal training data with a unique non-characteristic number; The historical load, the time data, environmental data and test data corresponding to the historical load are integrated into several training data groups and test data groups; the artificial intelligence model is trained by the several training data groups, and then the trained artificial intelligence model is tested by the test data group, and the artificial intelligence model is modified according to the test results to obtain a demand forecasting model whose input is the numbers corresponding to the time data, environmental data and classification data, and whose output is the predicted demand.

3. The heat control method based on cogeneration according to claim 1, characterized in that: The method of adjusting the heat supply of cogeneration according to the predicted demand includes: Count the current CHP supply; compare the forecast demand with the current CHP supply; When the predicted demand is less than the current CHP supply, the operating parameters of the CHP equipment are adjusted to reduce the heat supply of the CHP equipment; When the predicted demand is greater than the current CHP supply, the operating parameters of the CHP equipment are adjusted to increase the heat supply of the CHP equipment; When the predicted demand is equal to the current CHP supply, the operating parameters of the CHP equipment are maintained.

4. The heat control method based on cogeneration according to claim 1, characterized in that: The method of storing excess heat supply in a geothermal well comprises: The plate heat exchanger is started to exchange heat with the geothermal water in the geothermal well, and the geothermal water is re-injected into the geothermal well.

5. The heat control method based on cogeneration according to claim 1, characterized in that: The method of connecting the heat in the geothermal well to the peak-shaving boiler includes: The hot water in the geothermal well and the water source in the cogeneration are passed through the plate heat exchanger for heat exchange, and the geothermal water after heat exchange is sent to the evaporation side of the heat pump unit to extract heat; After heat is extracted, the geothermal water is reinjected into the geothermal well. The water from the outlet side of the plate heat exchanger and the condensing side of the heat pump unit is mixed and passed into the peak-shaving boiler for heating. The heated water is then sent to the water supply network of the cogeneration system.

Images