Cogeneration unit optimization operation adjustment method based on heat supply demand load prediction

Abstract

The invention discloses a cogeneration unit optimization operation adjustment method based on heat supply demand load prediction. The method comprises the steps of establishing a prediction model of a heat supply demand load at a future moment; predicting future short-term and long-term heat supply demand loads of the thermoelectric enterprises; constructing performance models of boilers and turbines of the whole plant, establishing constraint conditions and a target function, and performing load optimization distribution on the boilers and turbines of the whole plant; acquiring operation conditions and operation efficiency data of each boiler and steam turbine of the thermoelectric enterprise under different loads, and establishing a database; based on the distributed loads of all the boilers, the optimal operation working conditions of the boilers are searched in a database; the method comprises the following steps: collecting operation data of each boiler auxiliary equipment, and establishing a characteristic relation model between frequency and flow of each boiler auxiliary equipment; and according to the characteristic relation model, calculating the frequency of each boiler auxiliary machine corresponding to the optimal operation condition of the boiler. The heat supply pressure of the thermal power plant can be stable, equipment regulation and control are reduced, and the operation efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of cogeneration optimization control, and in particular relates to a method for optimizing the operation of a cogeneration unit based on heat demand load prediction. Background Art

[0002] Cogeneration units are usually composed of multiple boilers and steam turbines of various types. In addition to providing heat, they also need to generate electricity. The production system is relatively complex. There are still many problems and room for optimization and improvement in the production link on the boiler source side and the heating link on the heat network side.

[0003] For example, Chinese patent publication CN106437876A discloses a deep peak-shaving system and operating method for a combined heat and power (CHP) unit, which achieves power peak shaving and thermal load peak-shaving. Chinese patent publication CN110454764A discloses a thermoelectric decoupling system and operating method for a CHP unit, which achieves thermoelectric decoupling and recovers waste heat from steam turbine exhaust, resulting in high energy efficiency.

[0004] Boilers, a key piece of equipment in thermal power plants, burn coal. Their combustion mechanisms are complex, with various parameters interdependent and subject to varying time lags. The plant's DCS system generates tens of thousands of operational data points per second. This data contains a wealth of information on equipment operating status and lifespan, while also exhibiting complex interdependencies. Operators must manually assess the current status of this massive amount of data and respond quickly based on current and future needs, making it challenging to manage such a highly inert system like a thermal power plant boiler.

[0005] At the same time, considering that the technical level of thermal power plant operators is relatively weak and uneven, on the one hand, in daily operation, they often pay more attention to the control of parameters such as equipment safety, environmental protection and load changes, and ignore some important economic indicators; on the other hand, operators often form personal fixed adjustment habits after long-term operation, and it is difficult to break through fixed thinking and achieve optimized control of the entire cogeneration process. Summary of the Invention

[0006] The present invention provides a method for optimizing the operation of a cogeneration unit based on heat demand load prediction. The method can predict the heat demand load of a cogeneration enterprise in the future and guide the optimization of boiler and steam turbine operation, thereby achieving stable heat supply pressure in the cogeneration plant, reducing equipment control, and improving operation efficiency.

[0007] A method for optimizing the operation of a cogeneration unit based on heat demand load forecasting comprises the following steps:

[0008] (1) Obtain historical production data of thermal power companies and historical steam consumption data of corresponding heat users to establish a prediction model for future heating demand load;

[0009] (2) collecting real-time production data of thermal power enterprises and using the prediction model constructed in step (1) to predict the future short-term and long-term heating demand load of thermal power enterprises;

[0010] (3) Construct performance models for boilers and steam turbines throughout the plant, establish constraints and objective functions, and optimize load distribution for boilers and steam turbines throughout the plant based on the predicted heating demand load data;

[0011] (4) Obtain the operating conditions and efficiency data of each boiler and steam turbine of the thermal power enterprise under different loads through field tests and historical data mining, and establish a database through data cleaning and filling;

[0012] (5) Based on the loads of each boiler allocated in step (3), the optimal operating conditions of the boilers under the corresponding loads are screened step by step in the database of step (4);

[0013] (6) Collect the operating data of each boiler auxiliary equipment and steam turbine, establish the characteristic relationship model between the opening of the high-pressure regulating valve of the steam turbine and the flow rate, and the characteristic relationship model between the frequency and flow rate of each boiler auxiliary equipment, including the coal feeder frequency conversion characteristic model, the forced draft fan frequency conversion characteristic model and the induced draft fan frequency conversion characteristic model;

[0014] (7) adjusting the opening of the high-pressure regulating valve of the steam turbine based on the characteristic model between the opening of the high-pressure regulating valve of the steam turbine and the flow rate obtained in step (6) and based on the load of each steam turbine allocated in step (3);

[0015] At the same time, according to the characteristic relationship model between the equipment frequency and the flow rate obtained in step (6), the frequency of each boiler auxiliary equipment corresponding to the optimal boiler operating condition obtained in step (5) is calculated.

[0016] In step (1), a machine learning model is trained based on a recurrent neural network (RNN), a long short-term memory (LSTM), a gated recurrent unit (GRU), and spectrum analysis to predict the long-term heating demand load; a machine learning model is trained based on a long short-term memory (LSTM) to predict the short-term heating demand load;

[0017] Among them, the long-term heating demand load refers to the data after half an hour, and the short-term heating demand load refers to the data after half an hour.

[0018] The specific process of step (2) is:

[0019] (2-1) Call the measurement model constructed in step (1), input the operating data of the past half hour, and predict the short-term heating demand load in the next half hour and the long-term heating demand load in the next 6 hours;

[0020] (2-2) For the long-term heating demand load in the next 6 hours, the long-term forecast values of the first hour and the last 5 hours are read for curve fitting, and the forecast value of the first hour is used for interpolation to calculate the slope calculation value after interpolation. The short-term heating demand load is corrected according to the slope calculation value.

[0021] In step (3), the performance model of the boiler is the boiler efficiency model, and the formula is:

[0022] η=h(D b )

[0023] Where η is the boiler efficiency; D b is the steam output of the boiler; h(D b ) is the boiler coal consumption characteristic equation;

[0024] The performance model of the steam turbine is a multivariable model of the steam turbine inlet flow rate and the extraction and exhaust parameters, and a multivariable model of the steam turbine power generation and the extraction and exhaust parameters. The formula is:

[0025] D=f(D p ,D g )

[0026] P=g(D p ,D g )

[0027] Where, D is the steam inlet flow rate of the steam turbine; D p is the exhaust flow of the steam turbine; D g is the steam turbine medium pressure extraction steam heat flow rate; f(D p ,D g ) is the characteristic equation of steam inlet flow rate of the steam turbine; g(D p ,D g ) is the power generation characteristic of the steam turbine; P is the power generation power of the steam turbine.

[0028] In step (3), the constraints established are:

[0029] Unit capacity constraints:

[0030]

[0031] Extraction steam flow constraint:

[0032]

[0033] Exhaust steam flow constraint:

[0034]

[0035] Heating load equality constraint:

[0036]

[0037] Among them, D r is the total heat flow, D i is the steam inlet flow rate of the i-th unit; P i is the power generation capacity of the i-th unit; D gi is the extraction steam flow of the i-th unit, D pi is the exhaust flow rate of the i-th unit;

[0038] Boiler capacity constraints:

[0039]

[0040] Steam production equality constraint:

[0041]

[0042] Among them, (D i ) opt is the steam inlet flow after load optimization distribution of unit i, including turbine and desuperheater (if any); D′ is the main steam main pipe loss, D b The desuperheater and pressure reducer (double reduction) is used to make up for the missing steam when the turbine output is insufficient.

[0043] In step (3), the objective functions established based on the turbine side optimization objectives and the boiler side optimization objectives are specifically:

[0044] The optimization goal on the steam turbine side is to minimize the power generation steam consumption rate:

[0045]

[0046] Where d is the power generation steam consumption rate; m is the number of steam turbines; D i is the steam inlet flow rate of the i-th unit; P i is the power generation power of the i-th unit;

[0047] The optimization goal of the boiler side is to minimize the coal consumption rate for steam production:

[0048]

[0049] Where b is the coal consumption rate for steam production; n is the number of boilers; B i is the coal consumption of the i-th boiler, and the formula is:

[0050]

[0051] Among them, D bi is the steam production of the i-th boiler; H gi is the feed water enthalpy of the i-th boiler; H si is the main steam enthalpy of the i-th boiler; η i is the efficiency of the i-th boiler; Q is the calorific value of coal.

[0052] The specific process of step (5) is:

[0053] (5-1) Filter out the historical data set T1 with a higher efficiency than the current one in the database based on the boiler efficiency index;

[0054] (5-2) Select a similar dataset T2 with a similarity of more than 95% from the historical dataset T1 based on the matching parameters;

[0055] Among them, the matching parameters include boiler coal feed rate, boiler main steam temperature, boiler main steam pressure, lower calorific value of coal entering the furnace, received basis moisture, received basis ash;

[0056] (5-3) In the similar data set T2, the optimization intervals are screened in sequence according to the priority of the control parameters. The control parameters are divided into equally spaced intervals, and the interval with the highest frequency is screened. Then, within this interval, the intervals with the top 25% boiler efficiency are screened. Finally, the optimization result data set T3 is obtained by screening in sequence according to the priority. The priority order of the control parameters is furnace outlet oxygen content, primary and secondary air ratio, and bed temperature.

[0057] Among them, the control parameters include the oxygen content at the furnace outlet, the ratio of primary and secondary air, and the bed temperature; the current operating parameters include the current main steam flow rate and the current oxygen content at the furnace outlet;

[0058] (5-4) In the optimization result data set T3, the results are screened based on the similarity with the current operating parameters, and finally a set of results T4 with the highest similarity is obtained.

[0059] In step (6), the variable frequency characteristic model of the coal feeder is corrected by coupling the static frequency-flow relationship with the change amount at the next moment, specifically:

[0060] First, establish the static frequency-flow coupling relationship of the coal feeder:

[0061] F1=aD+b

[0062] Among them, F1 is the coal feeder frequency fitted according to the current coal feeder flow rate; D is the coal feeder flow rate; and a is the coefficient.

[0063] Then establish the coupling relationship between the coal feeder variation and frequency:

[0064] F2=z0+bΔD

[0065] Among them, F2 is the coal feeder frequency fitted according to the change value of the coal feeder flow rate; ΔD is the change value of the coal feeder flow rate; and b is the coefficient.

[0066] Finally, F1 and F2 are coupled twice using the Poly2D function:

[0067]

[0068] Among them, z0 is a constant term, and c, d, e, f, and g are all coefficients.

[0069] In step (6), the fan variable frequency characteristic model is obtained by establishing the coupling relationship between the fan frequency and the load and oxygen content:

[0070] F=z0+hΔO2+iΔD zq +F0

[0071] Among them, ΔO2 is the change in oxygen content at the furnace outlet; ΔD zq is the change value of the boiler main steam flow rate; F0 is the actual fan frequency at the current moment; h and i are coefficients.

[0072] In step (6), the variable frequency characteristic model of the induced draft fan is obtained by establishing a nonlinear surface model of the induced draft fan frequency and the standard state theoretical flue gas volume, the actual flue gas volume and the furnace negative pressure:

[0073] F=z0+jV y +kP

[0074] Among them, z0 is a constant term; V y is the actual flue gas volume; b is the furnace negative pressure coefficient; P is the target furnace negative pressure value; j and k are coefficients.

[0075] Compared with the prior art, the present invention has the following beneficial effects:

[0076] 1. This invention predicts the future load change trend of cogeneration enterprises based on a big data learning model, which can guide boiler operators to make load adjustments in advance, reduce pressure fluctuations in heating network pipelines, and ensure stable heating pressure;

[0077] 2. The present invention can realize real-time online calculation of boiler efficiency (update frequency 1 minute), and can intuitively understand the real-time operating status of the boiler;

[0078] 3. The present invention optimizes the operating conditions based on the boiler historical operation database accumulated by the boiler real-time efficiency calculation model, breaking through the original operating inertia thinking of boiler operators and improving boiler efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Figure 1This is a flow chart of a method for optimizing the operation of a cogeneration unit based on heat demand load forecasting according to an embodiment of the present invention. DETAILED DESCRIPTION

[0080] The present invention will be described in further detail below with reference to the accompanying drawings and examples. It should be noted that the following examples are intended to facilitate understanding of the present invention and do not have any limiting effect on the present invention.

[0081] like Figure 1 As shown, a method for optimizing the operation of a cogeneration unit based on heat demand load forecasting includes the following steps:

[0082] S1. Obtain historical steam usage data and related production data of thermal power company heat users, and train a load forecasting model for future long-term (24-hour) heating demand load based on recurrent neural networks (RNNs), long short-term memory networks (LSTMs), gated recurrent units (GRUs), and spectral analysis to obtain a long-term load forecasting model.

[0083] S2. Using date parameters such as the current hour, day, and month, as well as real-time operating parameters such as the heating main pipe flow, temperature, and pressure in the previous 30 minutes as input data, a long short-term memory network (LSTM) model is used to train a short-term (30-minute) heating demand load machine learning model to obtain a short-term load forecasting model.

[0084] S3. Call the model trained in step S1 and step S2, input the operating data of the previous 30 minutes as input parameters, and output the short-term heating demand load in the next half hour and the long-term heating demand load in the next 6 hours respectively.

[0085] S4. Correct the short-term heat demand load forecast data based on the long-term heat demand load forecast data, read the long-term forecast values of the previous hour and the next 5 hours for curve fitting, and use the forecast value of the previous hour for difference, calculate the slope calculation value after interpolation, and correct the short-term heat demand load forecast based on the slope calculation value.

[0086] S5. Through field tests and data coupling analysis, we obtain the multivariable model of each steam turbine's power generation and extraction and exhaust parameters, as well as the multivariable model of the steam inlet flow and extraction and exhaust parameters:

[0087] D=f(D p ,D g )

[0088] P=g(D p ,D g )

[0089] Through field tests, a load-boiler efficiency performance model for each boiler was established:

[0090] η=h(D b )

[0091] S6. Establish constraints based on boiler and steam turbine equipment performance and actual production conditions:

[0092] Unit capacity constraints:

[0093]

[0094] 5MW=P i min ≤P i ≤P i max =37.7MW

[0095] Extraction steam flow constraint:

[0096]

[0097] Exhaust steam flow constraint:

[0098]

[0099] Heating load equality constraint:

[0100]

[0101] Boiler capacity constraints:

[0102]

[0103] Steam production equality constraint:

[0104]

[0105] S7. Under the above constraint boundary conditions, establish the optimization allocation objective function for both the turbine and boiler sides respectively:

[0106] The optimization goal on the steam turbine side is to minimize the power generation steam consumption rate:

[0107]

[0108] The optimization goal of the boiler side is to minimize the coal consumption rate for steam production:

[0109]

[0110] S8. Under the premise of satisfying the predicted heating demand load calculated in step S2, calculation is performed according to the constraints of the above objective function and boundary condition function, and finally the load optimization distribution results of the boilers and turbines in the whole plant are obtained, as shown in Table 1.

[0111] Table 1 Results of optimized load distribution of boilers and turbines in the whole plant

[0112] Serial number category unit Numerical 1 Optimized load value of boiler No. 1 t / h 142.5 2 Optimized load value of boiler No. 2 t / h 188.6 3 Optimized load value of boiler #3 t / h 215.7 5 Optimized value of steam inlet flow of No. 1 turbine t / h 298.5 6 Optimized value of steam inlet flow of No. 2 turbine t / h 268.4

[0113] S9. Obtain the operating conditions and operating efficiency data of each boiler and steam turbine of the thermal power enterprise under different loads through field tests and historical data mining, and establish a database by cleaning and filling in the gaps in the data. The database contains the boiler efficiency of each operating condition calculated by the boiler efficiency calculation model;

[0114] "Matching parameters" include boiler coal feed rate, boiler main steam temperature, boiler main steam pressure, lower calorific value of incoming coal, received basis moisture, received basis ash; "control parameters" include furnace outlet oxygen content, primary and secondary air ratio, bed temperature; "current operating parameters" include current main steam flow, current furnace outlet oxygen content and other data;

[0115] S10. Based on the load of each device calculated in step S8, the database of step S9 is screened step by step to find the optimal operating condition of each device under the corresponding load, as shown in Table 2 below.

[0116] Table 2 Optimal operating conditions of boilers and steam turbines in the entire plant

[0117]

[0118]

[0119] S11. Collect the equipment operation data of each boiler auxiliary equipment (coal feeder, fan) and steam turbine, and establish the characteristic relationship model between the frequency and flow of boiler auxiliary equipment, as well as the characteristic model between the opening of the high-pressure regulating valve of the steam turbine and the flow.

[0120] S12. Based on the characteristic relationship model between the device frequency and the flow rate obtained in step S11, the device frequency corresponding to the optimal operating condition of each device in step S10 is calculated, as shown in Table 3 below.

[0121] Table 3 Recommended values for optimal operating frequency of boilers and turbines in the entire plant

[0122]

[0123]

[0124] The embodiments described above provide a detailed description of the technical solutions and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, supplements and equivalent substitutions made within the scope of the principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for optimizing the operation of a cogeneration unit based on heat demand load forecasting, characterized in that: The following steps are involved: (1) Obtain historical production data of thermal power companies and historical steam consumption data of corresponding heat users to establish a prediction model for future heating demand load; (2) collecting real-time production data of thermal power enterprises and using the prediction model constructed in step (1) to predict the future short-term and long-term heating demand load of thermal power enterprises; (3) Construct performance models for boilers and steam turbines throughout the plant, establish constraints and objective functions, and optimize load distribution for boilers and steam turbines throughout the plant based on the predicted heating demand load data; (4) Obtain the operating conditions and efficiency data of each boiler and turbine under different loads of thermal power enterprises through field tests and historical data mining, and establish a database through data cleaning and gap filling; (5) Based on the loads of each boiler allocated in step (3), the optimal operating conditions of the boilers under the corresponding loads are screened step by step in the database of step (4); (6) Collect the operating data of each boiler auxiliary equipment and steam turbine, establish the characteristic relationship model between the opening of the high-pressure regulating valve of the steam turbine and the flow rate, and the characteristic relationship model between the frequency and flow rate of each boiler auxiliary equipment, including the coal feeder frequency conversion characteristic model, the forced draft fan frequency conversion characteristic model and the induced draft fan frequency conversion characteristic model; (7) adjusting the opening of the high-pressure regulating valve of the steam turbine based on the characteristic model between the opening of the high-pressure regulating valve of the steam turbine and the flow rate obtained in step (6) and based on the load of each steam turbine allocated in step (3); At the same time, according to the characteristic relationship model between the equipment frequency and the flow rate obtained in step (6), the frequency of each boiler auxiliary equipment corresponding to the optimal boiler operating condition obtained in step (5) is calculated.

2. The method for optimizing operation and adjusting the combined heat and power unit based on heat demand load forecasting according to claim 1, characterized in that: In step (1), a machine learning model is trained based on a recurrent neural network (RNN), a long short-term memory (LSTM), a gated recurrent unit (GRU), and spectrum analysis to predict the long-term heating demand load; a machine learning model is trained based on a long short-term memory (LSTM) to predict the short-term heating demand load; Among them, the long-term heating demand load refers to the data after half an hour, and the short-term heating demand load refers to the data after half an hour.

3. The method for optimizing operation and adjusting the combined heat and power unit based on heat demand load forecasting according to claim 1, characterized in that: The specific process of step (2) is: (2-1) Call the measurement model constructed in step (1), input the operating data of the past half hour, and predict the short-term heating demand load in the next half hour and the long-term heating demand load in the next 6 hours; (2-2) For the long-term heating demand load in the next 6 hours, the long-term forecast values of the first hour and the last 5 hours are read for curve fitting, and the forecast value of the first hour is used for interpolation to calculate the slope calculation value after interpolation. The short-term heating demand load is corrected according to the slope calculation value.

4. The method for optimizing operation and adjusting cogeneration units based on heat demand load forecasting according to claim 1, characterized in that: In step (3), the performance model of the boiler is the boiler efficiency model, and the formula is: η=h(D b ) Where η is the boiler efficiency; D b is the steam output of the boiler; h(D b ) is the boiler coal consumption characteristic equation; The performance model of the steam turbine is a multivariable model of the steam turbine inlet flow rate and the extraction and exhaust parameters, and a multivariable model of the steam turbine power generation and the extraction and exhaust parameters. The formula is: D=f(D p ,D g ) P=g(D p ,D g ) Where, D is the steam inlet flow rate of the steam turbine; D p is the exhaust flow of the steam turbine; D g is the steam turbine medium pressure extraction steam heat flow rate; f(D p ,D g ) is the characteristic equation of steam inlet flow rate of the steam turbine; g(D p ,D g ) is a steam turbine P is the power generation capacity of the steam turbine.

5. The method for optimizing operation and adjusting cogeneration unit based on heat demand load forecasting according to claim 1, characterized in that: In step (3), the constraints established are: Unit capacity constraints: 5MW=P i min ≤P i ≤P i max =37.7MW Extraction steam flow constraint: Exhaust steam flow constraint: Heating load equality constraint: Among them, D r is the total heat flow, D i is the steam inlet flow rate of the i-th unit, P i is the power generation capacity of the i-th unit; D gi is the extraction steam flow of the i-th unit, D pi is the exhaust flow rate of the i-th unit; Boiler capacity constraints: Steam production equality constraint: Among them, (D i ) opt is the steam inlet flow rate after load optimization distribution of the i-th unit; D′ is the main steam main pipe loss, D b is the steam production of the boiler.

6. The method for optimizing operation and adjusting cogeneration units based on heat demand load forecasting according to claim 1, characterized in that: In step (3), the objective functions established based on the turbine side optimization objectives and the boiler side optimization objectives are specifically: The optimization goal on the steam turbine side is to minimize the power generation steam consumption rate: Where d is the power generation steam consumption rate; m is the number of steam turbines; D i is the steam inlet flow rate of the i-th unit; P i is the power generation power of the i-th unit; The optimization goal of the boiler side is to minimize the coal consumption rate for steam production: Where b is the coal consumption rate for steam production; n is the number of boilers; B i is the coal consumption of the i-th boiler, and the formula is: Among them, D bi is the steam production of the i-th boiler; H gi is the feed water enthalpy of the i-th boiler; H si is the main steam enthalpy of the i-th boiler; η i is the efficiency of the i-th boiler; Q is the calorific value of coal.

7. The method for optimizing operation and adjusting cogeneration units based on heat demand load forecasting according to claim 1, characterized in that: The specific process of step (5) is: (5-1) Filter out the historical data set T1 with a higher efficiency than the current one in the database based on the boiler efficiency index; (5-2) Select a similar dataset T2 with a similarity of more than 95% from the historical dataset T1 based on the matching parameters; Among them, the matching parameters include boiler coal feed rate, boiler main steam temperature, boiler main steam pressure, lower calorific value of coal entering the furnace, received basis moisture, received basis ash; (5-3) In the similar data set T2, the optimization intervals are screened in sequence according to the priority of the control parameters. The control parameters are divided into equally spaced intervals, and the interval with the highest frequency is screened. Then, within this interval, the intervals with the top 25% boiler efficiency are screened. Finally, the optimization result data set T3 is obtained by screening in sequence according to the priority. The priority order of the control parameters is furnace outlet oxygen content, primary and secondary air ratio, and bed temperature. Among them, the control parameters include the oxygen content at the furnace outlet, the ratio of primary and secondary air, and the bed temperature; the current operating parameters include the current main steam flow rate and the current oxygen content at the furnace outlet; (5-4) In the optimization result data set T3, the results are screened based on the similarity with the current operating parameters, and finally a set of results T4 with the highest similarity is obtained.

8. The method for optimizing operation and adjusting cogeneration units based on heat demand load forecasting according to claim 1, characterized in that: In step (6), the variable frequency characteristic model of the coal feeder is corrected by coupling the static frequency-flow relationship with the change amount at the next moment, specifically: First, establish the static frequency-flow coupling relationship of the coal feeder: F1=aD+b Wherein, F1 is the coal feeder frequency fitted according to the current coal feeder flow rate; D is the coal feeder flow rate; a is the coefficient; Then establish the coupling relationship between the coal feeder variation and frequency: F2=z0+bΔD Wherein, F2 is the coal feeder frequency fitted according to the change value of the coal feeder flow rate; ΔD is the change value of the coal feeder flow rate; b is the coefficient; Finally, F1 and F2 are coupled twice using the Poly2D function: Among them, z0 is a constant term, and c, d, e, f, and g are all coefficients.

9. The method for optimizing operation and adjusting cogeneration units based on heat demand load forecasting according to claim 1, characterized in that: In step (6), the fan variable frequency characteristic model is obtained by establishing the coupling relationship between the fan frequency and the load and oxygen content: F=z0+hΔO2+iΔD zq +F0 Among them, ΔO2 is the change in oxygen content at the furnace outlet; ΔD zq is the change value of the boiler main steam flow rate; F0 is the actual fan frequency at the current moment; h and i are coefficients.

10. The method for optimizing operation and adjusting cogeneration units based on heat demand load forecasting according to claim 1, characterized in that: In step (6), the variable frequency characteristic model of the induced draft fan is obtained by establishing a nonlinear surface model of the induced draft fan frequency and the standard state theoretical flue gas volume, the actual flue gas volume and the furnace negative pressure: F=z0+jV y +kP Among them, z0 is a constant term; V y is the actual flue gas volume; b is the furnace negative pressure coefficient; P is the target furnace negative pressure value; j and k are coefficients.

Images