Unit coordination control system and method for thermal power generating unit

Abstract

The invention provides a unit coordination control system and method for a thermal power generating unit, and relates to the technical field of thermal power generating unit coordination control. A prediction model is obtained through a historical load curve, and a real-time load curve is input into the prediction model to obtain a load prediction curve; according to the method, a standard power curve and a power increase rate curve of each single set of power generation equipment in the thermal power generating unit are obtained, a power control curve of each single set of power generation power is set, meanwhile, limiting conditions are set to be solved in MATLAB, cooperative control over the multiple single sets of power generation equipment in the thermal power generating unit is achieved, and meanwhile the requirements for load increase and increase rate are met; and according to the relationship between the power control curve and the load curve of each single set of power generation equipment, generating a thermal power generating unit state score, and setting a score threshold feedback result, thereby realizing safety guarantee of the continuous working state of the thermal power generating unit.

Description

Technical Field

[0001] The present invention relates to the technical field of coordinated control of thermal power units, and specifically to a unit coordinated control system and method for thermal power units. Background Art

[0002] As an important part of traditional power systems, the operating efficiency and stability of thermal power units directly affect the safety and economy of power supply. During actual operation, a thermal power unit contains multiple sets of power generation equipment. When supplying power to the power grid, multiple sets of power generation equipment work together. During the construction process of thermal power units, the models and performances of power generation equipment built at different times are different, and their power generation capabilities and response capabilities are also different. By coordinating and controlling the working conditions of each set of power generation equipment, the thermal power unit can meet the load demand of electricity.

[0003] In the prior art, thermal power units generally adopt the form of variable power supply, generating electricity at a lower power when the grid load is low, and increasing the power generation of thermal power units when the grid load increases. However, in actual applications, due to the coordination problems among multiple sets of power generation equipment inside the thermal power unit, and the power generation capabilities and response capabilities of each set of power generation equipment are different, there are challenges in the coordinated control of multiple sets of power generation equipment inside the thermal power unit when the grid load increases. Especially when the grid load increases sharply, the performance limitations of the thermal power unit itself easily lead to the power increase speed being less than the load increase speed, resulting in a power supply gap and affecting normal work and production.

[0004] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure, and thus it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present invention is to provide a unit coordinated control system and method for thermal power units to solve the problems raised in the above background art.

[0006] To achieve the above purpose, the present invention provides the following technical solutions: A unit coordinated control method for thermal power units, the specific steps include: Step 1: Obtain the time from startup to the maximum power of all single sets of power generation equipment in the thermal power unit, and select the longest time as the predicted window duration. Obtain the historical load curve of the thermal power unit, and input the load curve into the LSTM model for training to obtain a prediction model; Step 2: Collect the load curve during the operation of the thermal power unit in real time and input it into the prediction model for training, and obtain the load prediction curve within the predicted window duration through the trained prediction model; Step 3: Obtain the standard power curve of each single generating device in the thermal power unit. According to the standard power curve, obtain the power increase rate curve of each single generating device respectively. Set the power control curve of each single generating device within the prediction window duration and solve it in MATLAB. Set constraint I according to the load prediction curve, set constraint II according to the power increase rate curve, and set constraint III according to the minimum power requirement, and obtain the solved power control curve; Step 4: Obtain the highest load in the load curve and record the time when the highest load occurs. Obtain the highest control power and the time when the highest control power occurs through the power control curve. Construct the status score of the thermal power unit, set the score threshold, and feedback the result according to the relationship between the status score of the thermal power unit and the score threshold.

[0007] Furthermore, obtain the time from startup to the maximum power of all single generating devices in the thermal power unit, and select the longest time to calibrate the prediction window duration. Obtain the historical load curve of the thermal power unit, where the abscissa of the load curve is time and the ordinate is power.

[0008] Furthermore, divide the load curve into multiple curve segments with the same duration according to the prediction window duration, number all the curve segments in sequence and summarize them to form a training set. Set the (n + 1)-th curve segment as the label of the n-th curve segment, input it into the LSTM model for training, and calibrate the trained model as the prediction model.

[0009] Furthermore, collect the load curve of the thermal power unit during operation in real time and input it into the prediction model to obtain the load prediction curve. The abscissa of the load prediction curve is time, the time length is the length of the prediction window duration, and the ordinate is power. The load curve is updated in real time, and the latest load is added to the load curve.

[0010] Furthermore, obtain the standard power curve of each single generating device in the thermal power unit. The standard power curve is the curve of the power increase process of a single generating device from startup to the maximum generating power. The abscissa of the standard power curve is time, and the time span is from the 0 moment to the moment when the maximum generating power is reached. According to the standard power curve, obtain the standard power increase rate curve of each single generating device respectively. The abscissa of the standard power increase rate curve is power, and the ordinate is the power increase rate of this single generating device in this power state. The standard power increase rate curve represents the power increase rate of this single generating device in each power state.

[0011] Further, set the power control curve of each set of power generation equipment within the prediction window duration. The abscissa of the power control curve is time, and the ordinate is power. Solve the power control curve of each single set of power generation equipment in MATLAB, and set Constraint Ⅰ. The formula based on Constraint Ⅰ is as follows:

[0012] Wherein, is the power control curve of the th set of power generation equipment, which represents the power generation power of the th set of power generation equipment at moment, is the function of the power control curve of the th set of power generation equipment, is the safety factor, is the load prediction curve, which represents the load at moment here, is the power generation equipment retrieval variable, , , is the number of sets of power generation equipment; Set Constraint Ⅱ. The formula based on it is as follows:

[0013] Wherein, is the power increase rate of the power control curve, which represents the power increase rate of the th set of power generation equipment when the power generation power is , is the standard power increase rate curve, which represents the standard power increase rate of the th set of power generation equipment when the power is here; Set Constraint Ⅲ. The formula based on it is as follows:

[0014] Wherein, represents the minimization function, is the power control curve of the th set of power generation equipment, which represents the power of the th set of power generation equipment at moment here, is the current moment, is the last moment of the prediction window duration; Input Constraint Ⅰ, Constraint Ⅱ, and Constraint Ⅲ into MATLAB to obtain the power control curve of each set of power generation equipment after solution.

[0015] Further, the highest load is obtained from the load prediction curve on a daily basis and the time when the highest load occurs is recorded. The highest load is the numerical average of the highest loads in each period, and the highest load time is the average of the times when the highest loads occur in each period. According to the highest control power and the time when the highest control power occurs in the power control curve of each single set of power generation equipment, where the highest control power is the maximum value in the power control curve and the time when the highest control power occurs is the average of the times when the highest values occur in all power control curves, the status score of the thermal power unit is obtained. The formula is as follows:

[0016] Wherein, is the status score of the thermal power unit, is the time when the highest load occurs, is the time when the highest control power occurs, is the highest load, is the highest control power of the th set of power generation equipment, is the power generation equipment retrieval variable, , is the number of sets of power generation equipment, and are coefficients, , is the indication coefficient. When , is 1, and when , is 0.

[0017] Further, a scoring threshold is set, and the status score of the thermal power unit is compared and analyzed with the scoring threshold: When ,the system gives an alarm. When ,no alarm is given; The status score of the thermal power unit is obtained at a fixed frequency at intervals of half of the prediction window duration and analyzed with the scoring threshold.

[0018] The present invention further includes a unit coordinated control system for a thermal power unit. The system is used to execute the above-mentioned unit coordinated control method for a thermal power unit, and includes: A model construction module, which is used to obtain the time from startup to the maximum power of all single sets of power generation equipment in the thermal power unit, select the longest time and calibrate it as the prediction window duration, obtain the historical load curve of the thermal power unit, and input the load curve into the LSTM model for training to obtain a prediction model; The load acquisition module is used to collect the load curve during the operation of the thermal power unit in real time and input it into the prediction model for training, and obtain the load prediction curve within the prediction window duration through the trained prediction model; The power control solution module is used to obtain the standard power curve of each single power generation device in the thermal power unit, obtain the power increase rate curve of each single power generation device according to the standard power curve, set the power control curve of each single power generation device within the prediction window duration and solve it in MATLAB, set constraint I according to the load prediction curve, set constraint II according to the power increase rate curve, set constraint III according to the minimum power requirement, and obtain the solved power control curve; The state evaluation module is used to obtain the highest load in the load curve and record the occurrence time of the highest load, obtain the highest control power and the occurrence time of the highest control power through the power control curve, construct the state score of the thermal power unit, set the score threshold, and feedback the result according to the relationship between the state score of the thermal power unit and the score threshold.

[0019] Compared with the prior art, the beneficial effects of the present invention are: The present invention obtains the load prediction curve according to the prediction model, obtains the standard power curve and the power increase rate curve of each single device, and solves the power control curve of each single power generation device through the constraints, realizing the coordinated control of multiple single power generation devices in the thermal power unit, meeting the requirements of both load increase and increase rate at the same time, and forming the state score of the thermal power unit according to the power control curve and the load curve, realizing the safety guarantee of the continuous working state of the thermal power unit. Description of the Drawings

[0020] Figure 1 It is a schematic diagram of the overall method flow of the present invention; Figure 2 It is a schematic diagram of the system structure of the present invention. Detailed Embodiment

[0021] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to specific embodiments.

[0022] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should have the ordinary meanings understood by those with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar terms used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Words such as "including" or "comprising" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. Words such as "connected" or "linked" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right", etc. are only used to represent relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0023] Embodiment:

[0024] Please refer to Figure 1 , the present invention provides a technical solution: A unit coordinated control method for a thermal power unit, the specific steps include: Step 1: Obtain the time of all single sets of power generation equipment in the thermal power unit from startup to maximum power, select the longest time as the prediction window duration, obtain the historical load curve of the thermal power unit, and input the load curve into the LSTM model for training to obtain a prediction model; The said Step 1 includes the following content: Step 101: Obtain the time of all single sets of power generation equipment in the thermal power unit from startup to maximum power, select the longest time as the prediction window duration, obtain the historical load curve of the thermal power unit, where the abscissa of the load curve is time and the ordinate is power.

[0025] The core significance of obtaining the prediction window duration is that it physically limits the time coverage range of the prediction model, ensuring that the load prediction curve output by the prediction model can completely cover the longest time period required for all equipment startups. At the same time, collecting the historical load curve provides a structured time series data basis for subsequent model training. The load curve enables the prediction model to capture the dynamic change laws of demand over time, such as the differences between daytime and nighttime loads, seasonal fluctuations, etc., thereby enhancing the global adaptability of the prediction.

[0026] Step 102: Divide the load curve into multiple curve segments with the same duration according to the prediction window duration, number and summarize all the curve segments in sequence to form a training set, set the (n + 1)-th curve segment as the label of the n-th curve segment, input it into the LSTM model for training, and calibrate the trained model as the prediction model.

[0027] By segmenting the load model at equal time intervals, the content volume of the training set is increased. Meanwhile, using the load curve corresponding to each segmented curve as a label can improve the model accuracy. An LSTM model is used to train the load curve. Its advantage lies in being able to effectively capture the long-term dependencies in the time series. The prediction window duration is strictly limited to ensure that the prediction results match the physical capabilities of the actual power generation equipment. For example, if the model predicts that the demand will continue to rise in the next 3 hours, but a certain device takes 3 hours to reach full load, the demand curve output by the model will not include unreachable targets beyond this time window, thus avoiding the problem of the control strategy being out of touch with the device performance.

[0028] Step 2: Collect the load curve during the operation of the thermal power unit in real time and input it into the prediction model for training, and obtain the load prediction curve within the prediction window duration through the trained prediction model; The content of the said Step 2 includes the following: Collect the load curve during the operation of the thermal power unit in real time and input it into the prediction model to obtain the load prediction curve. The abscissa of the load prediction curve is time, and the time length is the length of the prediction window duration, and the ordinate is power. The load curve is updated in real time, and the latest load is added to the load curve.

[0029] Dynamically integrate the latest load curve to make the input data of the prediction model always reflect the current working conditions. This mechanism solves the deviation problem caused by data lag in the traditional static prediction model. For example, in the event of a sudden incident, real-time data update can quickly correct the prediction curve to ensure the timeliness and accuracy of subsequent control strategies.

[0030] Step 3: Obtain the standard power curve of each single power generation device in the thermal power unit, respectively obtain the power increase rate curve of each single power generation device according to the standard power curve, set the power control curve of each single power generation device within the prediction window duration and solve it in MATLAB. Set the constraint condition Ⅰ according to the load prediction curve, set the constraint condition Ⅱ according to the power increase rate curve, set the constraint condition Ⅲ according to the minimum power requirement, and obtain the solved power control curve; The content of the said Step 3 includes the following: Step 301: Obtain the standard power curve of each single generating device in the thermal power unit. The standard power curve is the curve of the power increase process of the single generating device from startup to the maximum generating power. The abscissa of the standard power curve is time, and the time span is from the 0 moment to the moment when the maximum generating power is reached. According to the standard power curve, obtain the standard power increase rate curve of each single generating device respectively. The abscissa of the standard power increase rate curve is power, and the ordinate is the power increase rate of this single generating device at this power state. The standard power increase rate curve represents the power increase rate of this single generating device at each power state.

[0031] Obtain the standard power curve and power increase rate curve of a single device, and clarify the dynamic performance boundary of each device. These data provide key constraint conditions for subsequent optimization. For example, the maximum power increase rate of the device limits the physical upper limit of power adjustment in the control strategy, avoiding equipment damage caused by overloading.

[0032] Step 302: Set the power control curve of each generating device within the prediction window duration. The abscissa of the power control curve is time, and the ordinate is power. Solve the power control curve of each single generating device in MATLAB, and set Constraint Ⅰ. The formula based on Constraint Ⅰ is as follows:

[0033] where, is the power control curve of the th generating device. Here, it represents the generating power of the th generating device at the moment. is the function of the power control curve of the th generating device. is the safety factor. is the load prediction curve. Here, it represents the load at the moment. is the generating device retrieval variable. , , is the number of sets of generating devices; The purpose of Constraint Ⅰ is to enable the generating power of the thermal power unit to always meet the load within the prediction window duration, avoiding the situation where the load increases while the generating power of the thermal power unit does not increase. The safety factor provides a safety redundancy for the power supply of the thermal power unit.

[0034] Set Constraint Ⅱ. The formula based on it is as follows:

[0035] where, is the power increase rate of the power control curve, indicating the The power generation equipment is The power increase rate when is the standard power increase rate curve, which represents the The power generation equipment is Standard power increase rate when In the face of increased load, the power control curve must also increase accordingly, but the increase in the power control curve is constrained by the single power generation equipment itself. The power increase rate of the power control curve does not exceed the standard power increase rate of each corresponding power generation equipment, so as to avoid overloading the single power generation equipment in order to increase power in a hurry. However, as the load increases sharply, every effort must be made to meet the load. When the load curve shows a sharp increase in power consumption demand after a specific power consumption node, the power increase rate of the thermal power unit at this node cannot be completed in a short time due to the physical limitations of its own equipment, and insufficient power supply will occur. Therefore, it is necessary to improve the power control curve of the thermal power unit in advance, so that when the load reaches the peak, the power generation power of the thermal power unit will also rise to meet the load, and adopt the "early bird flies first" approach to ensure the stability and timeliness of power supply.

[0036] Set constraint condition III, based on the following formula:

[0037] in, represents the minimization function, For the The power control curve of the power generation equipment is shown here. Power generation equipment in The power of the moment, For the current moment, The last moment of the prediction window length; On the basis of ensuring that the load curve and the load increase are met rapidly, the power control curve of the thermal power unit is minimized to avoid the situation where the thermal power unit is always operated at full power simply to meet the load, saving energy costs and maximizing benefits.

[0038] Input constraint conditions I, II and III into MATLAB to obtain the power control curve of each power generation equipment after solution.

[0039] The power control curve is set as a linear combination of basis functions, and constraints I, II and III are substituted into the coefficient equation group. Then the coefficients are solved by the least squares method, and the solved power control curve is brought in to calculate the residual norm for verification. If the residual norm condition is met, it is considered to be passed. If it does not pass, it is solved again.

[0040] Step 4: Obtain the highest load in the load curve and record the occurrence time of the highest load. Obtain the highest control power and the occurrence time of the highest control power through the power control curve. Construct the status score of the thermal power unit, set the score threshold, and feedback the result according to the relationship between the status score of the thermal power unit and the score threshold.

[0041] The said Step 4 includes the following contents: Step 401: Obtain the highest load from the load prediction curve on a daily basis and record the occurrence time of the highest load. The highest load is the numerical average of the highest loads in each cycle, and the highest load time is the average of the occurrence times of the highest loads in each cycle. According to the highest control power and the occurrence time of the highest control power of the power control curve of each single set of power generation equipment, the highest control power is the maximum value in the power control curve, and the occurrence time of the highest control power is the average of the occurrence times of the highest values in all power control curves. Obtain the status score of the thermal power unit, and the formula is as follows:

[0042] Wherein, is the status score of the thermal power unit, is the occurrence time of the highest load, is the occurrence time of the highest control power, is the highest load, is the highest control power of the th set of power generation equipment, is the power generation equipment retrieval variable, is the number of sets of power generation equipment, and are coefficients, is the indication coefficient. When , is 1. When , is 0.

[0043] Among them, by comparing the highest load and the highest control power, and the occurrence time of the highest load and the time when the highest control power is located, the state of the thermal power unit during the subsequent load increase process is judged. When the occurrence time of the highest load is after the time when the highest control power is located, the closer the time when the highest control power is located is to the occurrence time of the highest load, the more urgent the window period for dealing with the situation where power supply does not meet the load is, and the higher the state score of the thermal power unit is. When the occurrence time of the highest load is before or at the same time as the time when the highest control power is located, it means that the power generation power of the thermal power unit has met the peak load, and there is no need to deal with the problem that the power supply does not meet the load anymore, and the state score of the thermal power unit is directly set to 0. The sum of the maximum values of the power control curves of all single sets of power generation equipment within the prediction window duration represents the maximum power supply of the thermal power unit during the prediction window period. When the difference between the highest load and the maximum power supply of the thermal power unit is smaller, it means that the thermal power unit can almost meet the highest load, and the possibility of needing to deal with the situation where the power supply does not meet the load is smaller, and the state score of the thermal power unit is also smaller.

[0044] Step 402: Set the scoring threshold , and compare and analyze the state score of the thermal power unit with the scoring threshold: When occurs, the system gives an alarm. When occurs, no alarm is given; Obtain the state score of the thermal power unit at a fixed frequency at intervals of half of the prediction window duration and analyze it with the scoring threshold.

[0045] By comparing the scoring threshold, dynamically judge the working state of the generator set, and provide a pre-judgment basis for dealing with the sudden situation where the power supply does not meet the load. Recalculate the score regularly at intervals of half of the prediction window duration to ensure that the system can continuously adapt to the changes in working conditions. This mechanism enhances the robustness of the system and avoids the long-term judgment failure caused by misjudgment of a single score.

[0046] Please refer to Figure 2 , the present invention also provides a unit coordinated control system for a thermal power unit. The system is used to execute the above-mentioned unit coordinated control method for a thermal power unit, and includes: A model construction module, which is used to obtain the time from startup to the maximum power of all single sets of power generation equipment in the thermal power unit, and select the longest time to calibrate the prediction window duration, obtain the historical load curve of the thermal power unit, and input the load curve into the LSTM model for training to obtain a prediction model; A load acquisition module, which is used to collect the load curve during the operation of the thermal power unit in real time and input it into the prediction model for training, and obtain the load prediction curve within the prediction window duration through the trained prediction model; The power control solution module is used to obtain the standard power curve of each single power generation device in the thermal power unit, respectively obtain the power increase rate curve of each single power generation device according to the standard power curve, set the power control curve of each single power generation device within the prediction window duration and solve it in MATLAB, set constraint I according to the load prediction curve, set constraint II according to the power increase rate curve, set constraint III according to the minimum power requirement, and obtain the solved power control curve; The status evaluation module is used to obtain the highest load in the load curve and record the occurrence time of the highest load, obtain the highest control power and the occurrence time of the highest control power through the power control curve, construct the status score of the thermal power unit, set the score threshold, and feedback the result according to the relationship between the status score of the thermal power unit and the score threshold.

[0047] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to obtain a formula that is closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0048] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art can realize that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0049] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units. They may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0050] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all of them should be covered by the protection scope of this application.

Claims

1. A unit coordinated control method for a thermal power unit, characterized in that: The specific steps include: Step 1: Obtain the time from startup to maximum power of all single power generation equipment in the thermal power unit, select the longest time and calibrate it as the prediction window length, obtain the historical load curve of the thermal power unit, and input the load curve into the LSTM model for training to obtain the prediction model; Step 2: Collect the load curve of the thermal power unit in real time during operation and input it into the prediction model for training. The load prediction curve within the prediction window duration is obtained through the trained prediction model. Step 3: Obtain the standard power curve of each single power generation equipment in the thermal power unit, obtain the power increase rate curve of each single power generation equipment according to the standard power curve, set the power control curve of each single power generation equipment within the prediction window and solve it in MATLAB, set constraint condition I according to the load prediction curve, set constraint condition II according to the power increase rate curve, set constraint condition III according to the minimum power requirement, and obtain the solved power control curve; Step 4: Obtain the highest load in the load curve and record the time when the highest load occurs. Obtain the highest control power and the time when the highest control power occurs through the power control curve. Construct a status score for the thermal power unit, set a scoring threshold, and feedback the result based on the relationship between the status score of the thermal power unit and the scoring threshold.

2. A coordinated control method for a thermal power unit according to claim 1, characterized in that: The time from startup to maximum power of all single power generation equipment in the thermal power unit is obtained, and the longest time is selected to be calibrated as the prediction window length, and the historical load curve of the thermal power unit is obtained, where the horizontal axis of the load curve is time and the vertical axis is power.

3. A coordinated control method for a thermal power unit according to claim 2, characterized in that: The load curve is divided into multiple curve segments of the same length according to the prediction window duration. All curve segments are numbered in sequence and summarized to form a training set. The n+1th curve segment is set as the label of the nth curve segment, which is input into the LSTM model for training. The trained model is calibrated as the prediction model.

4. A coordinated control method for a thermal power unit according to claim 3, characterized in that: The load curve of the thermal power unit during operation is collected in real time and input into the prediction model to obtain the load prediction curve. The horizontal axis of the load prediction curve is time, the time length is the length of the prediction window, and the vertical axis is power. The load curve is updated in real time and the latest load is added to the load curve.

5. A coordinated control method for thermal power units according to claim 4, characterized in that: A standard power curve of each single power generation equipment in a thermal power unit is obtained, wherein the standard power curve is a curve of a power increase process of a single power generation equipment from startup to maximum power generation, wherein the horizontal coordinate of the standard power curve is time, and the time span is from time 0 to the time of maximum power generation, and a standard power increase rate curve of each single power generation equipment is obtained according to the standard power curve, wherein the horizontal coordinate of the standard power increase rate curve is power, and the vertical coordinate is the power increase rate of the single power generation equipment under this power state, and the standard power increase rate curve represents the power increase rate of the single power generation equipment under each power state.

6. A coordinated control method for thermal power units according to claim 5, characterized in that: Set the power control curve of each power generation equipment within the prediction window, where the horizontal axis of the power control curve is time and the vertical axis is power. Solve the power control curve of each single power generation equipment in MATLAB and set the restriction condition I. The restriction condition I is based on the following formula: ， in, For the The power control curve of the power generation equipment is shown here. Power generation equipment in The power generation at the moment, For the A function of the power control curve of a power generating equipment. is the safety factor, is the load forecast curve, which is represented here in The load of time, Retrieve variables for power generation equipment, , , is the number of sets of power generation equipment; Set the restriction condition II based on the following formula: ， in, is the power increase rate of the power control curve, indicating the The power generation equipment is The power increase rate when is the standard power increase rate curve, which represents the The power generation equipment is Standard power increase rate when Set constraint condition III, based on the following formula: ， in, represents the minimization function, For the The power control curve of the power generation equipment is shown here. Power generation equipment in The power of the moment, For the current moment, The last moment of the prediction window length; Input constraint conditions I, II and III into MATLAB to obtain the power control curve of each power generation equipment after solution.

7. A coordinated control method for thermal power units according to claim 6, characterized in that: The highest load is obtained from the load forecast curve on a daily basis and the time of occurrence of the highest load is recorded. The highest load is the numerical mean of the highest load in each cycle, and the time of the highest load is the average of the time when the highest load occurs in each cycle. According to the highest control power and the time of occurrence of the highest control power of the power control curve of each single power generation equipment, the highest control power is the maximum value of the value in the power control curve, and the time of occurrence of the highest control power is the average of the time of occurrence of the highest value in all power control curves, the state score of the thermal power unit is obtained, and the formula is as follows: ， in, Score the status of thermal power units. is the time when the maximum load occurs, is the time when the maximum control power occurs, For the highest load, For the The maximum control power of the power generation equipment, Retrieve variables for power generation equipment, , , is the number of power generation equipment, and is the coefficient, , is the indicator coefficient, when hour, is 1, when hour, is 0.

8. A coordinated control method for thermal power units according to claim 7, characterized in that: Setting scoring thresholds , compare and analyze the status score of the thermal power unit with the score threshold: when When When the police are not called; The status score of the thermal power unit is obtained at a fixed frequency with an interval of half the prediction window length and analyzed with the score threshold.

9. A unit coordination control system for a thermal power unit, the system being used to execute a unit coordination control method for a thermal power unit according to any one of claims 1 to 8, characterized in that: include: The model building module is used to obtain the time from startup to maximum power of all single power generation equipment in the thermal power unit, select the longest time to calibrate as the prediction window length, obtain the historical load curve of the thermal power unit, and input the load curve into the LSTM model for training to obtain the prediction model; The load acquisition module is used to collect the load curve of the thermal power unit in real time during operation and input it into the prediction model for training. The load prediction curve within the prediction window is obtained through the trained prediction model. The power control solution module is used to obtain the standard power curve of each single power generation equipment in the thermal power unit, obtain the power increase rate curve of each single power generation equipment according to the standard power curve, set the power control curve of each single power generation equipment within the prediction window and solve it in MATLAB, set the restriction condition I according to the load prediction curve, set the restriction condition II according to the power increase rate curve, set the restriction condition III according to the minimum power requirement, and obtain the solved power control curve; The status evaluation module is used to obtain the highest load in the load curve and record the time when the highest load occurs, obtain the highest control power and the time when the highest control power occurs through the power control curve, construct the status score of the thermal power unit, set the score threshold, and feedback the results based on the relationship between the status score of the thermal power unit and the score threshold.

Images