Decision-Making Method for Cold-End Energy Consumption Optimization of Thermal Power Units Based on Big Data Analysis

Abstract

This invention discloses a decision-making method for optimizing cold-end energy consumption in thermal power units based on big data analysis. This invention relates to the field of thermal power unit technology and solves the problem of disorganized parameter adjustments due to a lack of focus on core control variables and the failure to use cooling water circulation rate and cooling water inlet temperature as the sole control levers. This invention first performs a separate initial commissioning to lock in the commissioning characteristics, then combines the difference between the power generation efficiency and the preset range, matches weighting coefficients to solve for the optimal control characteristics, and performs precise tuning. This eliminates the need to adjust complex and redundant parameters, simplifying the control logic for cold-end cooling optimization. Simultaneously, by averaging multiple commissioning tests, it avoids the random errors of a single commissioning, significantly improving the reliability of the commissioning characteristics and realizing a vectorized and precise optimization of cooling parameters from empirically setpoint adjustments.

Description

Technical Field

[0001] This invention relates to the field of thermal power unit technology, specifically to a decision-making method for optimizing cold-end energy consumption of thermal power units based on big data analysis. Background Technology

[0002] As a core component affecting the power generation efficiency and energy consumption level of thermal power units, the proper matching of the cooling effect of the system directly determines the coal consumption for power supply and the power consumption for plant use. Currently, the operation and control of the cold end of thermal power units largely rely on the manual experience of maintenance personnel to perform setpoint operations, resulting in a relatively crude control method and numerous technical defects.

[0003] Traditional cold-end optimization and control often fail to focus on core control variables, neglecting to use cooling water circulation rate and cooling water inlet temperature as the sole control levers, resulting in chaotic parameter adjustments. Furthermore, it lacks a systematic process of first initial debugging to lock in characteristics and then quantitatively solving for the optimal control parameters, relying solely on empirical setpoints for adjustment. This makes it impossible to quantitatively match the optimal cooling parameters based on the unit's current operating conditions, leading to low debugging accuracy, high randomness, and difficulty in achieving precise optimization of cold-end energy consumption.

[0004] Furthermore, the lack of a real-time monitoring and abnormal fluctuation detection mechanism for unit power generation efficiency makes it impossible to promptly detect cooling matching deviations caused by unit load fluctuations, changes in environmental operating conditions, and equipment operating characteristic drift. Once power generation efficiency fluctuates significantly abnormally, it cannot automatically carry out closed-loop fine-tuning, which can easily lead to a gradual mismatch of the originally optimal cooling conditions. Under long-term operation, the cooling effect deviates from the economically reasonable range, resulting in serious waste of coal consumption and plant power.

[0005] In addition, traditional solutions lack standardized models for calculating power generation efficiency and determining operating conditions. They do not combine generator electrical parameters, main steam and turbine exhaust thermal parameters to solve for power generation efficiency in real time, nor do they set quantitative ranges to define operating conditions. The overall control logic is highly subjective and lacks versatility, making it difficult to adapt to the actual needs of thermal power units for adaptive optimization operation at the cold end under different loads and seasonal environments. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a decision-making method for optimizing the cold-end energy consumption of thermal power units based on big data analysis. This method solves the problems of not focusing on core control variables, not using cooling water circulation rate and cooling water inlet temperature as the sole control levers, and resulting in chaotic parameter adjustments.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a decision-making method for optimizing cold-end energy consumption of thermal power units based on big data analysis, comprising the following steps: Step 1: Based on the current coal consumption, confirm the associated preset power generation efficiency range. Simultaneously, based on the determined power generation efficiency, identify whether the cooling efficiency of the thermal power unit meets the standard at the current time, and generate a cooling adjustment signal based on the identification result. The specific method is as follows: Determine the generator line voltage U and generator line current I at the current moment, and use: P = The active power P of the generator unit is determined by ×U×I×cosφ, where φ is the preset generator power factor; Then, the main steam flow rate D and the main steam inlet enthalpy h associated with the thermal power unit are also considered. in and the enthalpy of steam turbine exhaust. out Monitoring was conducted to determine the method used: Confirm the power generation efficiency of the thermal power unit at the current moment; The specific method for confirming the cooling adjustment signal is as follows: Based on the current coal consumption, a preset power generation efficiency range is determined. This range is a preset range. If the power generation efficiency... If the power generation efficiency range is within a certain range, no processing is required; continuous monitoring is sufficient. If the power generation efficiency is less than the minimum value of the power generation efficiency range, it indicates that the power generation efficiency is too low, and a cooling increase signal is generated directly. If the power generation efficiency is greater than the maximum value of the power generation efficiency range, it indicates that the power generation efficiency is too high, and a cooling decrease signal is generated directly. Step 2: Based on the generated cooling debugging signal, perform initial debugging on the cooling water circulation rate and inlet water temperature associated with the cooling system, lock the debugging characteristics, and then determine the optimal characteristics and execute them based on the debugging characteristics and the difference values ​​associated with the power generation efficiency and the preset power generation efficiency range. The specific method for locking debug features is as follows: Based on the generated cooling debugging signal, execute the corresponding initial debugging process: Keep the original cooling water inlet temperature unchanged, and simultaneously adjust the cooling water circulation rate to a set of numerical units, which are preset units. After the adjustment is completed, execute a set of monitoring cycles, average the power generation efficiency associated with the monitoring cycle and record it as Fd, and confirm the efficiency difference associated with Fd and the original power generation efficiency. The efficiency difference is greater than 0, and the associated efficiency difference is recorded as the circulating water adjustment value. Reset the cooling water circulation rate and adjust the cooling water inlet temperature to a set of numerical units. After the adjustment is completed, execute another set of monitoring cycles and simultaneously confirm the efficiency difference associated with the monitoring cycle. Record the confirmed second set of efficiency differences as the inlet temperature adjustment value. Record the confirmed circulating water commissioning values ​​and inlet water temperature commissioning values ​​as commissioning characteristics; The specific method for determining the optimal feature is as follows: The power generation efficiency associated with the current moment is compared with the preset power generation efficiency range. The power generation efficiency is denoted as Fp, the minimum value of the preset power generation efficiency range is denoted as Fmin, and the maximum value is denoted as Fmax. The formula is: FJ = (Fmin + Fmax) ÷ 2. If the cooling adjustment signal is a cooling boost signal, then the following formula is used: Difference value = (FJ - Fp). If the cooling adjustment signal is a cooling reduction signal, then the following formula is used: Difference value = (Fp - FJ); Record the circulating water debugging value confirmed in the debugging features as TS1, and the confirmed inlet water temperature debugging value as TS2. Use: TS1×C1+TS2×C2=GL to lock the associated value GL, where C1 and C2 are preset fixed units and positive integers. Assign different unit values ​​to C1 and C2, execute several sets of value taking processes, and record the associated value GL corresponding to different value taking processes. Identify the selected values ​​associated with several correlation values ​​GL and difference values, where: selected value = |GL - difference value|, and determine the minimum value from the identified several different selected values, and denote C1 and C2 associated with the minimum value as the optimal features; The specific method for implementing optimal features is as follows: Based on the optimal characteristics, the cooling water circulation rate and inlet water temperature associated with the cooling system are adjusted. If the cooling adjustment signal is a cooling increase signal, the cooling water circulation rate is increased by C1 units and the cooling water inlet temperature is decreased by C2 units. If the cooling adjustment signal is a cooling decrease signal, the cooling water circulation rate is decreased by C1 units and the cooling water inlet temperature is increased by C2 units. Step 3: After the optimal features are determined and executed, the power generation efficiency of the thermal power unit is monitored in real time and the floating state is determined. If the floating state is abnormal, the fine-tuning process is executed to make the thermal power unit operate in a stable state. The specific method for determining floating state anomalies is as follows: The power generation efficiency monitored at the current moment is denoted as Fo, and the power generation efficiency monitored at the previous moment is denoted as Fn. It is determined whether the monitored Fo and Fn satisfy: |Fo-Fn|≥0.2×Fn. If they satisfy, it means that the floating state of the thermal power unit is abnormal. Otherwise, the power generation efficiency of the thermal power unit is continuously monitored. The specific method for executing the fine-tuning process is as follows: The system identifies whether the power generation efficiency is abnormally rising or falling. If it is abnormally rising, a cooling boosting method is used to increase the cooling water circulation rate and simultaneously reduce the cooling water inlet temperature until the monitored power generation efficiency Fo and the previous time Fn meet the judgment conditions. If the problem is due to a downward floating anomaly, a cooling reduction adjustment method will be adopted, which will reduce the cooling water circulation rate and simultaneously increase the cooling water inlet temperature until the monitored power generation efficiency Fo and the previous time Fn meet the judgment conditions.

[0008] This invention provides a decision-making method for optimizing cold-end energy consumption in thermal power units based on big data analysis. Compared with existing technologies, it has the following advantages: By first performing a separate initial test to lock in the test characteristics, and then combining the difference between the power generation efficiency and the preset range, the optimal control characteristics are solved by matching the weight coefficients and performing precise tuning. This eliminates the need to adjust complex and redundant parameters, simplifying the control logic of cold-end cooling optimization. At the same time, by taking the average of multiple test sets, the random errors of a single test are avoided, which greatly improves the reliability of the test characteristics and realizes the transformation of cooling parameters from empirical fixed value adjustment to vectorized precise optimization. After the optimal cooling parameters are executed, the fluctuation of the unit's power generation efficiency is continuously monitored in real time. A quantitative anomaly judgment threshold is set to identify abnormal operating conditions. Targeted closed-loop fine-tuning is carried out for upward and downward anomalies. This can promptly offset the cooling matching deviation caused by unit load fluctuations, changes in environmental operating conditions, and drift in equipment operating characteristics. This ensures that the unit always maintains stable operation within an economically reasonable power generation efficiency range, avoiding problems such as operating condition mismatch after optimization and deviation in cooling effect. Attached Figure Description

[0009] Figure 1 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0010] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0011] First Embodiment

[0012] Please see Figure 1 This application provides a decision-making method for optimizing cold-end energy consumption of thermal power units based on big data analysis, including the following steps: Step 1: Determine the coal consumption and power generation efficiency of the thermal power unit at the current moment, and based on the preset power generation efficiency range associated with the coal consumption, identify whether the cooling efficiency of the thermal power unit meets the standard at the current time, and generate a cooling adjustment signal based on the identification result. Specifically, the cooling adjustment signal includes a cooling increase signal or a cooling decrease signal. When the power generation efficiency is too low, it means that the corresponding cooling characteristic is too low, so it is necessary to appropriately increase the corresponding cooling characteristic. When the power generation efficiency is too high, it means that the corresponding cooling characteristic is too high, so it is necessary to appropriately decrease the corresponding cooling characteristic. The cooling characteristics include the circulation rate of cooling water and the inlet temperature of cooling water. Step 2: Based on the generated cooling debugging signal, perform initial debugging on the cooling water circulation rate and inlet water temperature associated with the cooling system, lock the debugging characteristics, and then determine the optimal characteristics and execute them based on the debugging characteristics and the difference values ​​associated with the power generation efficiency and the preset power generation efficiency range. Specifically, in the corresponding cooling debugging process, the cooling system generally only needs to debug the corresponding cooling water circulation rate and cooling water inlet water temperature. By executing the corresponding initial debugging process, the debugging characteristics associated with each unit value are identified. Subsequently, the optimal solution for cooling water circulation rate and cooling water inlet water temperature is confirmed based on the debugging characteristics, thereby determining the optimal characteristics and executing them. Step 3: After the optimal features are determined and executed, the power generation efficiency of the thermal power unit is monitored in real time, and the floating state is determined. If the floating state is abnormal, the fine-tuning process is executed to make the thermal power unit operate in a stable state.

[0013] Second Embodiment

[0014] In this embodiment, the main focus is on the confirmation process of the cooling debugging signal: The specific method for determining the power generation efficiency of a thermal power unit at the current moment is as follows: Determine the generator line voltage U and generator line current I at the current moment, and use: P = The active power P of the unit is determined by ×U×I×cosφ, where φ is the preset generator power factor (an angle value), which is determined in advance by the operator based on experience. Then, the main steam flow rate D and the main steam inlet enthalpy h associated with the thermal power unit are also considered. in and the enthalpy of steam turbine exhaust. out Monitoring was conducted to determine the method used: Confirm the power generation efficiency of the thermal power unit at the current moment; Based on the current coal consumption, a preset power generation efficiency range is determined. This range is a preset range, determined in advance by the operator based on experience. If the power generation efficiency... If the power generation efficiency range is within a certain range, no processing is required; continuous monitoring is sufficient. If the power generation efficiency is less than the minimum value of the power generation efficiency range, it indicates that the power generation efficiency is too low, and a cooling increase signal is generated directly. If the power generation efficiency is greater than the maximum value of the power generation efficiency range, it indicates that the power generation efficiency is too high, and a cooling decrease signal is generated directly. Specifically, when the power generation efficiency is too low, it means that the cooling capacity of the corresponding cooling system is low, resulting in low power generation efficiency of the corresponding thermal power unit. Therefore, it is necessary to appropriately increase the cooling capacity of the corresponding cooling system. When the power generation efficiency is too high, it means that the cooling capacity of the corresponding cooling system is high, resulting in high power generation efficiency of the corresponding thermal power unit. When the power generation efficiency is too high, it does not mean that the unit is operating optimally, but rather that there is excess cooling at the cold end. This will cause an increase in ineffective power consumption of the auxiliary equipment plant at the cold end. It is necessary to actively reduce cooling investment and reduce the output of circulating water and fans to keep the unit in an economically reasonable efficiency range.

[0015] Third Embodiment

[0016] In this embodiment, the main focus is on the process of confirming the optimal feature: The specific method for locking debugging features is as follows: Based on the generated cooling debugging signal, execute the corresponding initial debugging process: Keep the original cooling water inlet temperature unchanged, and simultaneously adjust the cooling water circulation rate to a set of numerical units (if the cooling adjustment signal is a cooling increase signal, adjust the cooling water circulation rate upward by a set of numerical units; if the cooling adjustment signal is a cooling decrease signal, adjust the cooling water circulation rate downward by a set of numerical units). The numerical units are preset units, which are determined in advance by the operator based on experience. After the adjustment is completed, a set of monitoring cycles is executed. The monitoring cycle is generally 1 minute. The power generation efficiency associated with the monitoring cycle is averaged and recorded as Fd. The efficiency difference associated with Fd and the original power generation efficiency is confirmed, and the efficiency difference is > 0. The associated efficiency difference is recorded as the circulating water adjustment value. Reset the cooling water circulation rate and adjust the cooling water inlet temperature to a set of numerical units (if the cooling adjustment signal is a cooling increase signal, lower the cooling water inlet temperature by a set of numerical units; if the cooling adjustment signal is a cooling decrease signal, raise the cooling water inlet temperature by a set of numerical units). After the adjustment is completed, execute another set of monitoring cycles and simultaneously confirm the efficiency difference associated with the monitoring cycle. Record the confirmed second set of efficiency differences as the inlet temperature adjustment value. The confirmed circulating water adjustment value and inlet water temperature adjustment value are recorded as adjustment features. Specifically, as a further embodiment of this embodiment, multiple adjustment processes can be executed. After the efficiency difference associated with the first adjustment process is determined, the adjustment process is carried out again under the original value state, and the corresponding efficiency difference is confirmed. The process is stopped after adjusting five sets of values. The efficiency difference associated with the five adjustment processes is averaged, and the confirmed average value is recorded as the corresponding circulating water adjustment value or inlet water temperature adjustment value, thereby locking the corresponding adjustment feature. Based on the confirmed debugging characteristics, the specific method for determining and executing the optimal characteristics is as follows: The power generation efficiency associated with the current moment is compared with the preset power generation efficiency range. The power generation efficiency is denoted as Fp, the minimum value of the preset power generation efficiency range is denoted as Fmin, and the maximum value is denoted as Fmax. The formula is: FJ = (Fmin + Fmax) ÷ 2. If the cooling adjustment signal is a cooling boost signal, then the following formula is used: Difference value = (FJ - Fp). If the cooling adjustment signal is a cooling reduction signal, then the following formula is used: Difference value = (Fp - FJ); Record the circulating water debugging value confirmed in the debugging features as TS1, and the confirmed inlet water temperature debugging value as TS2. Use: TS1×C1+TS2×C2=GL to lock the associated value GL, where C1 and C2 are preset fixed units and positive integers. Assign different unit values ​​to C1 and C2, execute several sets of value taking processes, and record the associated value GL corresponding to different value taking processes. Identify the selected values ​​associated with several correlation values ​​GL and difference values, where: selected value = |GL - difference value|, and determine the minimum value from the identified several different selected values, and denote C1 and C2 associated with the minimum value as the optimal features; Based on the optimal characteristics, the cooling water circulation rate and inlet water temperature associated with the cooling system are adjusted. If the cooling adjustment signal is a cooling increase signal, the cooling water circulation rate is increased by C1 units and the cooling water inlet temperature is decreased by C2 units. If the cooling adjustment signal is a cooling decrease signal, the cooling water circulation rate is decreased by C1 units and the cooling water inlet temperature is increased by C2 units. Specifically, after determining the debugging characteristics associated with the cooling water circulation rate and inlet water temperature, the difference between the current power generation efficiency and the corresponding preset range is identified. Then, different numerical units are matched through the debugging characteristics to lock the corresponding selected value. Based on the proximity state associated with the selected value and the difference, the closest state is determined. Since the assigned values ​​of C1 and C2 are different, the associated proximity states are also different. From the value selection process of C1 and C2, the optimal process is locked to achieve the optimal value selection effect. Then, the process is executed to ensure the overall debugging effect of the comprehensive debugging.

[0017] Fourth embodiment

[0018] In this embodiment, compared to the above embodiments, the main focus is on the fine-tuning process for handling abnormal floating states: The specific method for determining floating state anomalies is as follows: The power generation efficiency monitored at the current moment is denoted as Fo, and the power generation efficiency monitored at the previous moment is denoted as Fn. It is determined whether the monitored Fo and Fn satisfy: |Fo-Fn|≥0.2×Fn. If they satisfy, it means that the floating state of the thermal power unit is abnormal. Otherwise, the power generation efficiency of the thermal power unit is continuously monitored. The specific method for executing the fine-tuning process is as follows: The system identifies whether the power generation efficiency is abnormally rising or falling. If it is abnormally rising, a cooling boosting method is used to increase the cooling water circulation rate and simultaneously reduce the cooling water inlet temperature until the monitored power generation efficiency Fo and the previous time Fn meet the judgment conditions. If the abnormality is due to a downward floating, a cooling reduction adjustment method will be adopted, which will reduce the cooling water circulation rate and simultaneously increase the cooling water inlet temperature until the monitored power generation efficiency Fo and the previous time Fn meet the judgment conditions. Specifically, if the upward movement is abnormal, the cooling process will be adjusted upward to appropriately reduce the corresponding power generation efficiency; if the downward movement is abnormal, the cooling process will be adjusted downward to increase the corresponding power generation efficiency.

[0019] Some of the data in the above formulas are numerical calculations with dimensions removed, and the contents not described in detail in this specification are all prior art known to those skilled in the art.

[0020] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A decision-making method for optimizing cold-end energy consumption of thermal power units based on big data analysis, characterized in that, Includes the following steps: Step 1: Based on the current coal consumption, confirm the associated preset power generation efficiency range, and simultaneously identify whether the cooling efficiency of the thermal power unit meets the standard based on the determined power generation efficiency, and generate a cooling adjustment signal based on the identification result. Step 2: Based on the generated cooling debugging signal, perform initial debugging on the cooling water circulation rate and inlet water temperature associated with the cooling system, lock the debugging characteristics, and then determine the optimal characteristics and execute them based on the debugging characteristics and the difference values ​​associated with the power generation efficiency and the preset power generation efficiency range. Step 3: After the optimal features are determined and executed, the power generation efficiency of the thermal power unit is monitored in real time, and the floating state is determined. If the floating state is abnormal, the fine-tuning process is executed to make the thermal power unit operate in a stable state.

2. The decision-making method for cold-end energy consumption optimization of thermal power units based on big data analysis according to claim 1, characterized in that, In step one, the specific method for determining the power generation efficiency is as follows: Determine the generator line voltage U and generator line current I at the current moment, and use: P = The active power P of the generator unit is determined by ×U×I×cosφ, where φ is the preset generator power factor; Then, the main steam flow rate D and the main steam inlet enthalpy h associated with the thermal power unit are also considered. in and the enthalpy of steam turbine exhaust. out Monitoring was conducted to determine the method used: Confirm the power generation efficiency of the thermal power unit at the current moment.

3. The decision-making method for cold-end energy consumption optimization of thermal power units based on big data analysis according to claim 2, characterized in that, In step one, the specific method for confirming the cooling adjustment signal is as follows: Based on the current coal consumption, a preset power generation efficiency range is determined. This range is a preset range. If the power generation efficiency... If the power generation efficiency is within a certain range, no processing is required; continuous monitoring is sufficient. If the power generation efficiency is less than the minimum value of the power generation efficiency range, it indicates that the power generation efficiency is too low, and a cooling increase signal is generated directly. If the power generation efficiency is greater than the maximum value of the power generation efficiency range, it indicates that the power generation efficiency is too high, and a cooling decrease signal is generated directly.

4. The decision-making method for cold-end energy consumption optimization of thermal power units based on big data analysis according to claim 1, characterized in that, In step two, the specific method for locking the debugging features is as follows: Based on the generated cooling debugging signal, execute the corresponding initial debugging process: Keep the original cooling water inlet temperature unchanged, and simultaneously adjust the cooling water circulation rate to a set of numerical units, which are preset units. After the adjustment is completed, execute a set of monitoring cycles, average the power generation efficiency associated with the monitoring cycle and record it as Fd, and confirm the efficiency difference associated with Fd and the original power generation efficiency. The efficiency difference is greater than 0, and the associated efficiency difference is recorded as the circulating water adjustment value. Reset the cooling water circulation rate and adjust the cooling water inlet temperature to a set of numerical units. After the adjustment is completed, execute another set of monitoring cycles and simultaneously confirm the efficiency difference associated with the monitoring cycle. Record the confirmed second set of efficiency differences as the inlet temperature adjustment value. Record the confirmed circulating water commissioning values ​​and inlet water temperature commissioning values ​​as commissioning characteristics.

5. The decision-making method for cold-end energy consumption optimization of thermal power units based on big data analysis according to claim 4, characterized in that, In step two, the specific method for determining the optimal feature is as follows: The power generation efficiency associated with the current moment is compared with the preset power generation efficiency range. The power generation efficiency is denoted as Fp, the minimum value of the preset power generation efficiency range is denoted as Fmin, and the maximum value is denoted as Fmax. The formula is: FJ = (Fmin + Fmax) ÷ 2. If the cooling adjustment signal is a cooling boost signal, then the following formula is used: Difference value = (FJ - Fp). If the cooling adjustment signal is a cooling reduction signal, then the following formula is used: Difference value = (Fp - FJ); Record the circulating water debugging value confirmed in the debugging features as TS1, and the confirmed inlet water temperature debugging value as TS2. Use: TS1×C1+TS2×C2=GL to lock the associated value GL, where C1 and C2 are preset fixed units and positive integers. Assign different unit values ​​to C1 and C2, execute several sets of value taking processes, and record the associated value GL corresponding to different value taking processes. Identify the selected values ​​associated with several correlation values ​​GL and difference values, where: selected value = |GL - difference value|, and determine the minimum value from the identified several different selected values, and denote C1 and C2 associated with the minimum value as the optimal features.

6. The decision-making method for cold-end energy consumption optimization of thermal power units based on big data analysis according to claim 5, characterized in that, In step two, the specific method for implementing the optimal feature is as follows: Based on the optimal characteristics, the cooling water circulation rate and inlet water temperature associated with the cooling system are adjusted. If the cooling adjustment signal is a cooling increase signal, the cooling water circulation rate is increased by C1 units and the cooling water inlet temperature is decreased by C2 units. If the cooling adjustment signal is a cooling decrease signal, the cooling water circulation rate is decreased by C1 units and the cooling water inlet temperature is increased by C2 units.

7. The decision-making method for cold-end energy consumption optimization of thermal power units based on big data analysis according to claim 1, characterized in that, In step three, the specific method for determining the abnormal floating state is as follows: The power generation efficiency monitored at the current moment is denoted as Fo, and the power generation efficiency monitored at the previous moment is denoted as Fn. It is determined whether the monitored Fo and Fn satisfy: |Fo-Fn|≥0.2×Fn. If they satisfy, it means that the thermal power unit is in an abnormal floating state. Otherwise, the power generation efficiency of the thermal power unit is continuously monitored.

8. The decision-making method for cold-end energy consumption optimization of thermal power units based on big data analysis according to claim 7, characterized in that, In step three, the specific method for performing the fine-tuning process is as follows: The system identifies whether the power generation efficiency is abnormally rising or falling. If it is abnormally rising, a cooling boosting method is used to increase the cooling water circulation rate and simultaneously reduce the cooling water inlet temperature until the monitored power generation efficiency Fo and the previous time Fn meet the judgment conditions. If the problem is due to a downward floating anomaly, a cooling reduction adjustment method will be adopted, which will reduce the cooling water circulation rate and simultaneously increase the cooling water inlet temperature until the monitored power generation efficiency Fo and the previous time Fn meet the judgment conditions.

Images