Thermal power generating unit controllable parameter safe operation and energy consumption online evaluation system
By constructing a thermodynamic model of thermal power units and performing correlation analysis of controllable parameters, dynamic priority division, and directional data acquisition, the complex parameter coupling problem of increased energy consumption in thermal power units was solved, and efficient energy consumption monitoring and control were achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-01-12
- Publication Date
- 2026-04-07
AI Technical Summary
In the monitoring and handling of increased energy consumption in thermal power units, the parameters are highly coupled and the root cause is difficult to locate. Traditional detection methods are inefficient and have a serious lag in response, making it difficult to achieve early detection and early intervention.
A thermodynamic model of the unit is constructed. The correlation state is divided and the correlation score is calculated through the controllable parameter correlation analysis module. Dynamic priority division is carried out, and the controllable parameter directional acquisition module is used to mark the control order. The model is then evaluated in conjunction with the safety performance assessment module.
It enables efficient control of controllable parameters of thermal power units, reduces time lag and energy loss caused by blind experimentation, and improves the efficiency of early identification and handling of energy consumption anomalies.
Smart Images
Figure CN121810122A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of thermal power unit operation evaluation technology, and more specifically, to a system for online evaluation of the safe operation and energy consumption of controllable parameters of thermal power units. Background Technology
[0002] Thermal power generation, as a pillar energy source in the current power system, directly impacts national energy costs and environmental protection through its operational economics. With the deepening of energy structure adjustments and power market reforms, thermal power units face severe challenges in achieving deep peak shaving and energy conservation. In actual unit operation, coal consumption for power generation (i.e., energy consumption) is the core indicator for measuring its economic efficiency. However, an abnormal increase in coal consumption is not caused by a single factor, but rather by a complex interplay of multiple controllable parameters deviating from their optimal setpoints.
[0003] Currently, the industry generally faces the following technical challenges in monitoring and addressing the increase in energy consumption of thermal power units: 1. High complexity of parameter coupling, making root cause localization difficult. A unit has dozens of controllable parameters, covering multiple systems such as combustion, steam-water systems, and auxiliary equipment. These parameters are strongly coupled; degradation of one parameter often masks or triggers anomalies in another. Traditional operational analysis relies on the experience of operators, monitoring and adjusting individual parameters in isolation, making it difficult to identify dominant degradation combinations under multi-parameter coupling.
[0004] 2. Traditional detection methods are crude, inefficient, and severely delayed in response. While existing operation monitoring systems can collect and display data for all parameters, their analysis model is essentially a "comprehensive and uniform" inspection. When energy consumption shows an upward trend, operation or analysis personnel need to compare and investigate the historical curves of all relevant controllable parameters one by one. This method is labor-intensive, time-consuming, and highly dependent on personal experience. From the discovery of abnormal coal consumption to the manual completion of data analysis and identification of possible causes, it often takes several hours or even longer. During this period, uneconomical operating conditions persist, leading to the continuous accumulation of additional energy losses, making "early detection and early intervention" impossible.
[0005] To address the aforementioned issues, there is an urgent need for an online assessment system for the safe operation and energy consumption of controllable parameters in thermal power units that can perform parameter correlation analysis. Summary of the Invention
[0006] The purpose of this invention is to provide a system for safe operation of controllable parameters and online energy consumption assessment of thermal power units, so as to solve the problems mentioned in the background art.
[0007] To achieve the above objectives, a system for online assessment of safe operation and energy consumption of controllable parameters of thermal power units is provided, including a unit thermodynamic model construction module, a controllable parameter correlation analysis module, a controllable parameter directional acquisition module, and a safety performance assessment module. The unit thermodynamic model construction module is used to collect historical monitoring data, obtain ideal benchmark values of energy consumption corresponding to different boundary conditions, locate response parameters under different energy consumption increase trends, and construct the unit thermodynamic model. The controllable parameter correlation analysis module is used to classify the correlation status between response parameters and controllable parameters under different energy consumption increase trends, calculate the correlation score between individual controllable parameters and combined controllable parameters, and perform dynamic priority classification. The controllable parameter directional acquisition module performs adjustment priority marking according to the priority of individual controllable parameters and combined controllable parameters, and obtains the corresponding controllable parameter values according to the adjustment priority. The safety performance evaluation module combines the final controllable parameter values with historical monitoring data to evaluate the safety performance of thermal power units.
[0008] As a further improvement to this technical solution, the method for constructing the unit thermodynamic model in the unit thermodynamic model construction module includes the following steps: S101. Based on the design specifications and operating history of thermal power units, the numerical range of each boundary condition is divided and discretized. S102. Eliminate combinations that do not exist under physical constraints, and bind the numerical range combinations of each boundary condition to construct a valid set of numerical range combinations; S103. Based on historical operational big data, select the historical lowest energy consumption value corresponding to different boundary condition value range combinations in the numerical range combination set, and set it as the ideal benchmark value under this working condition. S104. Establish the mapping relationship between the combination of boundary condition numerical ranges, ideal reference values, energy consumption increase, and response parameters.
[0009] As a further improvement to this technical solution, the method for dividing the correlation state in the controllable parameter correlation analysis module (20) includes the following steps: S201. Collect the response trend and numerical variation range of the response parameters, and construct a set of trend data. ,in Indicates a response trend. This indicates the range of numerical variation, and t represents the data collection period. S202. Obtain the initial values of the controllable parameters corresponding to the set of changing trends. And adjust the values Construct a set of controllable parameter changes ,in For the initial state tolerance, Represents the initial set of values. Indicates the adjustment tolerance. Represents the set of control values; S203. Establish the mapping relationship between the set of changing trends and the set of changes in controllable parameters.
[0010] As a further improvement to this technical solution, the method for calculating the correlation score of a single controllable parameter in the controllable parameter correlation analysis module (20) includes the following steps: S2010. Extract the numerical variation range of each response parameter. ; S2011, under the statistical historical control, each conforms to the corresponding initial value And adjust the values Effective number of times controllable parameters are adjusted and total number of regulation ; S2012, Calculate the controllability of the controllable parameter. , representing the correlation score of a single controllable parameter; S2013. Establish a control rate threshold and mark controllable parameters whose control rates exceed the control rate threshold as undetermined controllable parameters. As a further improvement to this technical solution, the method for calculating the correlation score of combined controllable parameters in the controllable parameter correlation analysis module (20) includes the following steps: S2014. Combine the undetermined controllable parameters matched in the response parameters to obtain various combined controllable parameters; S2015. Calculate the correlation score of combined controllable parameters under the subsequent control order of a single undetermined controllable parameter. S2016, Calculate the correlation score of combined controllable parameters. ,in - , where represents the control rate of a single undetermined controllable parameter in the combination, and n represents the number of single undetermined controllable parameters in the combination; S2017. Combining the correlation scores of single undetermined controllable parameters and combined controllable parameters, construct the numerical variation range of each response parameter. The corresponding set of undetermined controllable parameters is associated and mapped, and dynamic priority is assigned.
[0011] As a further improvement to this technical solution, the correlation score of the undetermined controllable parameter in S2017 is positively correlated with the dynamic priority.
[0012] As a further improvement to this technical solution, the method for adjusting the order marking in the controllable parameter directional acquisition module (30) includes the following steps: S301. Real-time monitoring of the actual values of various response parameters; when an increase in energy consumption is detected. S302. Combine the set of changing trends to perform actual numerical analysis on each response parameter and obtain the set of actual values corresponding to the unit time period. S303. Extract the corresponding actual numerical range and compare it with the numerical range in the trend set. Perform range matching; S304. Select the response parameters that successfully match the range, index them to the corresponding controllable parameter change set, collect the values of the undetermined controllable parameters for the controllable parameter change set, and extract the actual initial values from the initial value set. The controllable parameters are used as valid undetermined controllable parameters; S305. For the selected valid undetermined controllable parameters, based on the correlation scores of the specific undetermined controllable parameters and the combined controllable parameters, the remaining undetermined controllable parameters are prioritized as individual undetermined controllable parameters and combined undetermined controllable parameters. S306. Based on priority, perform control sequence planning for single undetermined controllable parameters and combinations of undetermined controllable parameters.
[0013] As a further improvement to this technical solution, the control order planning in S306 adopts a depth-first traversal selection method. The specific planning method steps are as follows: S3061. Select undetermined controllable parameters that conform to the current response parameters, and determine the control rate of the corresponding undetermined controllable parameters. The first phase of regulatory priority division is conducted to construct the main regulatory sequence; S3062. Numerical control is performed on the high-priority undetermined controllable parameters in the main control sequence according to the control order. S3063. If the adjustment of a single parameter fails to restore energy consumption to the baseline, then obtain the combined undetermined controllable parameters of the current undetermined controllable parameters, calculate the combined correlation score, and perform secondary sequential adjustment based on the score. S3064. When all combinations of the current undetermined controllable parameters have been adjusted in order and energy consumption has not yet recovered, the next undetermined controllable parameter in the order of adjustment shall be adjusted according to the results of the first adjustment order division. S3065. Repeat the above steps to complete the adjustment or energy consumption recovery of all pending controllable parameters.
[0014] As a further improvement to this technical solution, the method for evaluating the safety performance of thermal power units in the safety performance evaluation module (40) includes the following steps: S401. Obtain the actual values of each response parameter before adjustment and the actual values of each response parameter after adjustment, and calculate the absolute value of the difference between the two as the adjustment range difference. S402. Collect the standard values of various controllable parameters corresponding to the ideal baseline value of the current energy consumption, compare them with the actual values of the corresponding controllable parameters after adjustment, and calculate the absolute value of the difference between the two as the residual deviation. S403. Adopt the 3σ principle to set multi-level safety assessment standards based on historical steady-state monitoring data, including safe operation threshold, early warning threshold and extreme risk threshold; S404. Calculate the comprehensive risk index by combining the adjustment range difference and residual deviation, and compare and evaluate it with the multi-level safety assessment standards.
[0015] As a further improvement to this technical solution, the formula for calculating the comprehensive risk index is as follows: ; in As an ideal reference value, and For safety weighting coefficients, To adjust the difference in amplitude, This is the residual deviation; The RI value of the comprehensive risk index is negatively correlated with the safety performance of thermal power units, and when the RI value exceeds the preset limit risk threshold, it is judged as unqualified.
[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: In this online assessment system for the safe operation and energy consumption of controllable parameters in thermal power units, the controllable parameter correlation analysis module divides the correlation between response parameters and controllable parameters under different energy consumption increase trends, obtains the numerical change relationship between response parameters and controllable parameters, and calculates the correlation score between individual controllable parameters and combinations of controllable parameters. Combined with the controllable parameter directional acquisition module, the system prioritizes the controllable parameters for regulation. By matching the numerical monitoring results of response parameters with associated controllable parameters, it can eliminate controllable parameters with low correlation in advance and regulate controllable parameters sequentially according to the regulation order. This ensures the efficiency of controllable parameter regulation and avoids the regulation of irrelevant controllable parameters. Attached Figure Description
[0017] Figure 1 This is a block diagram of the overall system structure of the present invention.
[0018] The meanings of the labels in the diagram are as follows: 10. Unit thermodynamic model construction module; 20. Controllable parameter correlation analysis module; 30. Controllable parameter directional acquisition module; 40. Safety performance assessment module. Detailed Implementation
[0019] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples.
[0020] Please see Figure 1 As shown, a system for online evaluation of safe operation and energy consumption of controllable parameters of thermal power units is provided, including a unit thermodynamic model construction module 10, a controllable parameter correlation analysis module 20, a controllable parameter directional acquisition module 30, and a safety performance evaluation module 40. The unit thermodynamic model construction module 10 is connected to the historical database of the distributed control system (DCS) of the thermal power unit. It is used to collect historical monitoring data information, obtain the ideal benchmark value of energy consumption corresponding to different boundary conditions, locate the response parameters under different energy consumption increase trends, and construct the unit thermodynamic model. The controllable parameter correlation analysis module 20, based on big data statistical analysis technology, is used to classify the correlation status between response parameters and controllable parameters under different energy consumption increase trends, calculate the correlation score between individual controllable parameters and combined controllable parameters, and perform dynamic priority classification. The controllable parameter directional acquisition module 30, as the system's execution scheduling unit, marks the control order according to the priority of individual controllable parameters and combined controllable parameters, and obtains the corresponding controllable parameter values according to the control order. The safety performance evaluation module 40 is used to perform closed-loop verification of the system state after regulation, and to evaluate the safety performance of the thermal power unit by combining the final controllable parameter values with historical monitoring data.
[0021] The details are as follows: First, to ensure the consistency of benchmarks in subsequent evaluations, it is necessary to establish a database of ideal benchmark values for energy consumption under various given boundary conditions. Boundary conditions are defined as objective parameters that cannot be actively adjusted by operators during unit operation due to grid dispatch instructions or natural environmental constraints, such as load instructions, ambient temperature, and the type of combustible material (coal). Based on thermodynamic principles and unit design specifications, different combinations of boundary conditions necessarily correspond to a theoretically lowest energy consumption state, i.e., the ideal benchmark value under that state. If the real-time monitored energy consumption data significantly exceeds the aforementioned ideal benchmark value, it indicates that the operating parameters under the current state (i.e., controllable parameters such as air volume and feedwater pressure) deviate from the optimal operating condition, requiring timely evaluation and adjustment.
[0022] Based on the above problems, this solution uses the unit thermodynamic model construction module 10 to collect and analyze historical monitoring data. It employs a working condition division and historical data filtering method to obtain ideal baseline values of energy consumption corresponding to different boundary conditions, locate response parameters under different energy consumption increase trends, and construct a unit thermodynamic model to obtain the influence relationship between boundary conditions, ideal baseline values of energy consumption, and response parameters. The specific construction method is as follows: S101. Based on the design specifications and operating history of thermal power units, the numerical range of each boundary condition is divided (discretized). For example, for ambient temperature, multiple ambient temperature ranges are divided according to a step size of 5℃ or 10℃; for unit load, load ranges are divided according to a step size of 5% or 10% of the rated load.
[0023] S102. The discretized boundary adjustment numerical range is divided and processed by association. Considering the physical constraints of unit operation and actual scheduling scenarios (e.g., the cooling water temperature in winter is unlikely to occur under high-temperature conditions in summer), combinations that do not exist physically are eliminated, and an effective set of combinations of construction numerical ranges is constructed.
[0024] S103. Based on historical operational big data, select the historical lowest energy consumption value corresponding to different boundary condition value range combinations in the numerical range combination set, set it as the ideal benchmark value under this operating condition, and at the same time, statistically analyze the distribution range of key operating parameters corresponding to this ideal operating condition.
[0025] S104. Input the ideal baseline value into the numerical range combination set for binding. When the actual operation of the unit deviates from the ideal state, it will induce changes in the values of different response parameters, such as a significant increase in flue gas temperature, a deviation of the oxygen content in the flue gas from the optimal value, and an increase in the carbon content of fly ash / slag. These are specific manifestations induced by increased energy consumption. Obtain the response trend (numerical change) and the range of numerical change of each response parameter as an external manifestation of increased energy consumption, and form an evaluation route: boundary condition numerical range combination - ideal baseline value - energy consumption increase - response parameter, construct the unit thermodynamic model, and use it as a basis for tracing subsequent increases in energy consumption.
[0026] Furthermore, since there are many reasons for increased energy consumption, improper adjustment of the corresponding controllable parameters will also change with the increase in energy consumption. Although controllable parameters can be adjusted, which provides a solution for subsequent energy consumption control, the changing trends and ranges of different controllable parameters are different. If all controllable parameters are traversed for detection and evaluation, the detection efficiency will be greatly affected. Therefore, it is necessary to use the controllable parameter correlation analysis module 20 to divide the correlation state between the response parameters and controllable parameters under different energy consumption increase trends.
[0027] Its core logic lies in establishing a historical experience database between fault phenomena (response parameters) and adjustment measures (controllable parameters). The specific construction process is as follows: S201. Collect the response trend and numerical variation range of the response parameters, and construct a set of trend data. ,in This indicates a trend in response (e.g., monotonically increasing, monotonically decreasing, or oscillating). This indicates the range of numerical changes during the abnormal period, and > t represents the data collection period, meaning that within the data collection period t, the highest value of the current response parameter is... The lowest value is .
[0028] S202. Based on historical operation records, trace the controllable parameter operation behavior that led to the above response trend, and obtain the initial value of the corresponding controllable parameter. And adjust the values Adjusting the values It is the difference obtained by analyzing historical successful control cases, that is, in the actual control process, for different initial values The values are adjusted accordingly, and the corresponding energy consumption status is acquired in real time until the energy consumption returns to the ideal baseline value. At this point, the difference between the initial controllable parameter value and the adjusted (normal) controllable parameter value is the adjustment value. .
[0029] S203. Introduce a tolerance factor to construct a set of controllable parameter variations. ,in The initial state tolerance (taken as 1-2 times the statistical standard deviation or 3 times the instrument accuracy). This represents the initial set of values, meaning that the initial values of controllable parameters falling within this range are considered potential causes of increased energy consumption of the same type. Adjustment tolerance (value is ±10% of the target adjustment amount). This represents the set of controllable values; any adjustment of controllable parameters within this range is considered an effective improvement measure.
[0030] S204. Establish a set of changing trends With controllable parameter change set The mapping relationship between them serves as the associated state between response parameters and controllable parameters.
[0031] Furthermore, since most of the response parameters resulting from increased energy consumption are caused by changes in the combination of multiple controllable parameters (i.e., multivariate coupling effect), it is necessary to consider not only the impact of a single controllable parameter on the response parameter, but also the impact of the combination of multiple controllable parameters on the response parameter. Based on this, after completing the correlation analysis between the response parameter and the controllable parameter, the correlation score between a single controllable parameter and the combination of controllable parameters is calculated by combining the corresponding set of change trends and the set of controllable parameter changes. The specific calculation method is as follows: S301, Calculation of Single-Item Correlation Score: Extract the range of numerical variation of each response parameter. Obtain the corresponding initial values for each condition under historical regulation. And adjust the values The amount of adjustment of the controllable parameters.
[0032] Calculate the controllability of the controllable parameter. : Statistics show the range of changes in response parameters within historical data. At that time, the number of times a certain controllable parameter is effectively regulated. And the total number of adjustments to all controllable parameters under this operating condition. .
[0033] The formula for calculating the control rate is: ; This indicator represents the frequency percentage of times that a controllable parameter has solved such problems in historical operations, which is equivalent to the correlation score of a single controllable parameter.
[0034] S302, Parameter Filtering: Set control rate threshold (For example, set to 0.1% or 10%), the control rate The controllable parameters are marked as undetermined controllable parameters.
[0035] Based on the controllability of the undetermined controllable parameter The control priority of a single undetermined controllable parameter is divided, i.e., the control rate. The higher the value, the earlier the subsequent adjustment priority of the corresponding single undetermined controllable parameter.
[0036] S303, Calculation of Combinatorial Correlation Score: For a combination of undetermined controllable parameters, the correlation score of the combined controllable parameters is calculated under the subsequent control order of a single undetermined controllable parameter. Its physical significance lies in evaluating the effectiveness of the coordinated regulation of multiple parameters.
[0037] Corresponding combined correlation score The calculation formula is: ; in - , where represents the control rate of a single undetermined controllable parameter in the combination, and n represents the number of single undetermined controllable parameters in the combination; That is, the control rate of a single undetermined controllable parameter in the combination. The higher the score, the higher the corresponding composite correlation score. The larger.
[0038] S304, Dynamic Priority Sort: During the specific comparison process, the comparison between combined controllable parameters needs to ensure that the number of combinations of combined controllable parameters (i.e., the value of n) is consistent. For example, in a certain response parameter, the specific undetermined controllable parameters include... The corresponding regulatory priority is > > > .
[0039] Assuming that due to actual process limitations, only the following combinations are considered: and , and , and as well as and .
[0040] According to the product rule: because > and > Then there must be and .
[0041] In the specific combinations mentioned above, and The combination with the highest correlation score corresponds to the highest regulatory priority. and The combination of these has the lowest correlation score and the lowest regulatory order.
[0042] When there are combinations of identical undetermined controllable parameters, calculate the control rates among the individual undetermined controllable parameters after the combination. The product is used as the correlation score for the combined undetermined controllable parameters.
[0043] By combining the correlation scores of individual undetermined controllable parameters and combined controllable parameters, the numerical variation range of each response parameter is constructed. The corresponding set of undetermined controllable parameters is associated and mapped, and dynamic priority is assigned. That is, the association score of the undetermined controllable parameters is positively correlated with the dynamic priority.
[0044] Furthermore, to further ensure the efficiency of assessing subsequent increases in energy consumption, the controllable parameter directional acquisition module 30 performs controllability priority marking based on the priority of individual and combined controllable parameters. This step transforms the theoretical correlation priority into an operational sequence that the actual control system can execute. The specific controllability priority marking process is as follows: S401, Real-time Status Matching: The system monitors the actual values of various response parameters in real time. When the system detects an event that triggers an increase in energy consumption, it analyzes the actual values of various response parameters in conjunction with the aforementioned set of trends.
[0045] Obtain the set of actual values within a unit time period, extract the corresponding range of actual values, and compare it with the range of value changes in the trend set. Perform range matching, that is, determine whether the current actual value range falls within the corresponding value change range. middle.
[0046] S402, Preliminary screening of controllable parameters: Select the response parameters that successfully match the range, and index them to the corresponding set of controllable parameter changes through the mapping relationship.
[0047] Data acquisition of undetermined controllable parameters for the set of controllable parameter changes: Real-time reading of controllable parameter values in the current DCS system, and extraction of the actual initial values within the initial value set. The controllable parameters are used as valid undetermined controllable parameters.
[0048] S403, Priority and Order Generation: By combining the specific effective undetermined controllable parameters with the correlation scores of the previously calculated single / combined controllable parameters, the selected undetermined controllable parameters are prioritized.
[0049] Based on priority, the control sequence planning is performed for single and combined undetermined controllable parameters, and a traversal selection method is used for sequence planning. The specific planning method is as follows: First, select the controllable parameters that match the current response parameters to generate a control sequence. This embodiment uses a depth-first traversal strategy of "single parameter priority, combined parameter progression," and the specific steps are as follows: S4031, Master sequence construction: Based on the regulation rate of the corresponding undetermined controllable parameter. The first regulatory priority division (i.e., the main regulatory queue) is carried out.
[0050] S4032, Single-parameter control attempt: Based on the control priority, select high-priority undetermined controllable parameters in the main queue in sequence to issue numerical control commands.
[0051] S4033, Combined Branch Traversal: If the single parameter control does not achieve the expected effect (i.e., energy consumption does not return to the baseline), the combined control logic is triggered.
[0052] Obtain the combination of undetermined controllable parameters that are strongly correlated with the corresponding undetermined controllable parameters, that is: other parameters that meet the initial value set and are valid combinations with the current parameter.
[0053] Calculate the association score after combination Based on the score, the system iterates through all combinations of the currently undetermined controllable parameters and performs secondary sequential control.
[0054] S4034. Backtracking and Iteration: After all combinations of currently undetermined controllable parameters have been adjusted in order (or energy consumption has returned to normal), if the problem is still not solved, the next undetermined controllable parameter in the order of adjustment is selected for adjustment based on the results of the first adjustment.
[0055] Repeat the above steps until all undetermined controllable parameters are adjusted or the system energy consumption is restored to within the safe threshold.
[0056] Finally, the safety performance of the thermal power unit is evaluated by combining the final controllable parameter values with historical monitoring data through the safety performance evaluation module 40.
[0057] Specifically, the system obtains the actual values of each response parameter before adjustment. and the actual values of each adjusted response parameter. The absolute value of the difference between the two is calculated as the adjustment range difference. ; At the same time, standard values of various controllable parameters corresponding to the ideal energy consumption baseline value under the current boundary conditions are collected. and the adjusted actual value Compare the two and calculate the absolute value of the difference as the residual bias. .
[0058] Based on this, the system has preset a multi-level safety assessment standard that includes a safe operation threshold, an early warning threshold, and an extreme risk threshold, and quantitatively classifies and judges the above differences.
[0059] To address the issue of ambiguous standards in existing technologies, the specific method for setting multi-level security assessment standards in this embodiment is as follows: The 3σ principle was used to statistically analyze the collected historical steady-state monitoring data and construct a normal distribution model: Safe operating threshold (green zone): set to ±1σ (approximately ±5%) of the ideal baseline value; Warning threshold (yellow zone): set to ±1σ~±3σ (approximately ±5%~±15%). Extreme risk threshold (red zone): set to exceed ±3σ (approximately equal to ±15%), or directly retrieve the design limit parameters in the manufacturer's manual of the thermal power unit as this threshold.
[0060] Subsequently, the comprehensive risk index (RI) is calculated using the following logic: ; in, and These are the process normalized deviation rate and the result normalized deviation rate, respectively. and These are the corresponding safety weight coefficients (e.g., α=0.4, β=0.6, depending on the emphasis on stability). The larger the RI value, the higher the safety risk.
[0061] Let's take the oxygen content in flue gas as an example for a specific explanation: Assuming the ideal baseline value of flue gas oxygen content under current operating conditions The threshold is set at 3.0%, and the exponential line for the early warning threshold is set at 0.15. Weighting coefficients are set at α=0.5 and β=0.5.
[0062] If the oxygen content in the flue gas was 5.0% before adjustment, it will drop to 3.2% after system adjustment.
[0063] At this point, the adjustment range difference is =∣5.0%−3.2%∣=1.8%, residual deviation is =|3.2%−3.0%|=0.2%.
[0064] If we only look at the results (traditional method): the deviation rate = 0.2 / 3.0 ≈ 0.067, which looks very close to the ideal value and is easily misjudged as absolutely safe.
[0065] This system evaluation (introducing a process dimension): Process deviation rate = 1.8 / 3.0 = 0.6.
[0066] The overall risk index RI = 0.5 × 0.6 + 0.5 × 0.067 = 0.3335.
[0067] Judgment result: The calculated result of 0.3335 far exceeds the warning threshold of 0.15. This physically indicates that although the final oxygen content was adjusted, the system has just undergone a drastic correction (from 5.0 to 3.2), which means that the combustion conditions are extremely unstable and the air supply system may be over-regulated, so a warning must be issued.
[0068] According to the assessment criteria, when the oxygen content is too high, it means that the air supply volume is too large, which leads to increased heat loss from flue gas and increased power consumption of the fans; when the oxygen content is too low, it will lead to incomplete combustion of fuel, producing CO and causing coking or corrosion in the reducing atmosphere, which will also affect the current operation of thermal power plants and create safety hazards.
[0069] In summary, this system quantifies the abstract concept of operational stability by introducing the adjustment range difference and the comprehensive risk index. When the RI value approaches the extreme risk threshold, it indicates that although the current response parameters may be close to the standard in numerical terms, the underlying risk of system oscillation is extremely high, thus accurately identifying hidden faults where the numerical values meet the standards but the operating conditions are unstable.
[0070] In summary, this invention uses the controllable parameter correlation analysis module 20 to divide the correlation state between response parameters and controllable parameters under different energy consumption increase trends, constructs a mapping relationship between energy consumption characterization and adjustment methods, and calculates the correlation score between individual controllable parameters and combined controllable parameters.
[0071] This process, in conjunction with the controllable parameter directional acquisition module 30, dynamically divides the controllable parameter adjustment order, which is essentially a decoupling and dimensionality reduction process for the complex multivariable coupled system of thermal power units.
[0072] In actual operation, the system matches the controllable parameters associated with the numerical monitoring results of the response parameters, and can exclude controllable parameters with low correlation (i.e., correlation scores below the preset threshold) in advance, thereby greatly reducing the control search space.
[0073] Ultimately, by adjusting controllable parameters sequentially according to the optimized control order, the time lag and energy loss caused by blind trial and error can be significantly reduced, ensuring the efficiency of controllable parameter adjustment and completely avoiding ineffective control of irrelevant controllable parameters.
Claims
1. A system for online assessment of controllable parameters and energy consumption of thermal power units, characterized in that: It includes a unit thermodynamic model construction module (10), a controllable parameter correlation analysis module (20), a controllable parameter directional acquisition module (30), and a safety performance evaluation module (40); The unit thermodynamic model construction module (10) is used to collect historical monitoring data, obtain the ideal benchmark value of energy consumption corresponding to different boundary conditions, locate the response parameters under different energy consumption increase trends, and construct the unit thermodynamic model. The controllable parameter correlation analysis module (20) is used to classify the correlation status between response parameters and controllable parameters under different energy consumption increase trends, calculate the correlation score between single controllable parameters and combined controllable parameters, and perform dynamic priority classification. The controllable parameter directional acquisition module (30) performs adjustment priority marking according to the priority of single controllable parameters and combined controllable parameters, and obtains the corresponding controllable parameter values according to the adjustment priority; The safety performance evaluation module (40) combines the final controllable parameter values with historical monitoring data to evaluate the safety performance of the thermal power unit.
2. The online assessment system for controllable parameters and energy consumption of thermal power units according to claim 1, characterized in that: The method for constructing the unit thermodynamic model in the unit thermodynamic model construction module (10) includes the following steps: S101. Based on the design specifications and operating history of thermal power units, the numerical range of each boundary condition is divided and discretized. S102. Eliminate combinations that do not exist under physical constraints, and bind the numerical range combinations of each boundary condition to construct a valid set of numerical range combinations; S103. Based on historical operational big data, select the historical lowest energy consumption value corresponding to different boundary condition value range combinations in the numerical range combination set, and set it as the ideal benchmark value under this working condition. S104. Establish the mapping relationship between the combination of boundary condition numerical ranges, ideal reference values, energy consumption increase, and response parameters.
3. The online assessment system for controllable parameters and energy consumption of thermal power units according to claim 1, characterized in that: The method for dividing the correlation state in the controllable parameter correlation analysis module (20) includes the following steps: S201. Collect the response trend and numerical variation range of the response parameters, and construct a set of trend data. ,in Indicates a response trend. This indicates the range of numerical variation, and t represents the data collection period. S202. Obtain the initial values of the controllable parameters corresponding to the set of changing trends. And adjust the values Construct a set of controllable parameter changes ,in For the initial state tolerance, Represents the initial set of values. Indicates the adjustment tolerance. Represents the set of control values; S203. Establish the mapping relationship between the set of changing trends and the set of changes in controllable parameters.
4. The online assessment system for controllable parameters and energy consumption of thermal power units according to claim 3, characterized in that: The method for calculating the correlation score of a single controllable parameter in the controllable parameter correlation analysis module (20) includes the following steps: S2010. Extract the numerical variation range of each response parameter. ; S2011, under the statistical historical control, each conforms to the corresponding initial value And adjust the values Effective number of times controllable parameters are adjusted and total number of regulation ; S2012, Calculate the controllability rate of controllable parameters. , representing the correlation score of a single controllable parameter; S2013. Establish a control rate threshold and mark controllable parameters whose control rates exceed the control rate threshold as undetermined controllable parameters.
5. The online assessment system for controllable parameters and energy consumption of thermal power units according to claim 4, characterized in that: The method for calculating the correlation score of combined controllable parameters in the controllable parameter correlation analysis module (20) includes the following steps: S2014. Combine the undetermined controllable parameters matched in the response parameters to obtain various combined controllable parameters; S2015. Calculate the correlation score of combined controllable parameters under the subsequent control order of a single undetermined controllable parameter. S2016, Calculate the correlation score of combined controllable parameters. ,in - , where represents the control rate of a single undetermined controllable parameter in the combination, and n represents the number of single undetermined controllable parameters in the combination; S2017. Combining the correlation scores of single undetermined controllable parameters and combined controllable parameters, construct the numerical variation range of each response parameter. The corresponding set of undetermined controllable parameters is associated and mapped, and dynamic priority is assigned.
6. The online assessment system for controllable parameters and energy consumption of thermal power units according to claim 5, characterized in that: The correlation score of the undetermined controllable parameter in S2017 is positively correlated with the dynamic priority.
7. The online assessment system for controllable parameters and energy consumption of thermal power units according to claim 1, characterized in that: The method for adjusting the order marking in the controllable parameter directional acquisition module (30) includes the following steps: S301. Real-time monitoring of the actual values of various response parameters; when an increase in energy consumption is detected. S302. Combine the set of changing trends to perform actual numerical analysis on each response parameter and obtain the set of actual values corresponding to the unit time period. S303. Extract the corresponding actual numerical range and compare it with the numerical range in the trend set. Perform range matching; S304. Select the response parameters that successfully match the range, index them to the corresponding set of controllable parameter changes, collect the values of the undetermined controllable parameters for the set of controllable parameter changes, and extract the actual initial values from the initial value set. The controllable parameters are used as valid undetermined controllable parameters; S305. For the selected valid undetermined controllable parameters, based on the correlation scores of the specific undetermined controllable parameters and the combined controllable parameters, the remaining undetermined controllable parameters are prioritized as individual undetermined controllable parameters and combined undetermined controllable parameters. S306. Based on priority, perform control sequence planning for single undetermined controllable parameters and combinations of undetermined controllable parameters.
8. The online assessment system for controllable parameters and energy consumption of thermal power units according to claim 7, characterized in that: The control order planning in S306 adopts a depth-first traversal selection method. The specific planning steps are as follows: S3061. Select undetermined controllable parameters that conform to the current response parameters, and determine the control rate of the corresponding undetermined controllable parameters. The first phase of regulatory priority division is conducted to construct the main regulatory sequence; S3062. Numerical control is performed on the high-priority undetermined controllable parameters in the main control sequence according to the control order. S3063. If the adjustment of a single parameter fails to restore energy consumption to the baseline, then obtain the combined undetermined controllable parameters of the current undetermined controllable parameters, calculate the combined correlation score, and perform secondary sequential adjustment based on the score. S3064. When all combinations of the current undetermined controllable parameters have been adjusted in order and energy consumption has not yet recovered, the next undetermined controllable parameter in the order of adjustment shall be adjusted according to the results of the first adjustment order division. S3065. Repeat the above steps to complete the adjustment or energy consumption recovery of all pending controllable parameters.
9. The online assessment system for controllable parameters and energy consumption of thermal power units according to claim 1, characterized in that: The method for evaluating the safety performance of thermal power units in the safety performance evaluation module (40) includes the following steps: S401. Obtain the actual values of each response parameter before adjustment and the actual values of each response parameter after adjustment, and calculate the absolute value of the difference between the two as the adjustment range difference. S402. Collect the standard values of various controllable parameters corresponding to the ideal baseline value of the current energy consumption, compare them with the actual values of the corresponding controllable parameters after adjustment, and calculate the absolute value of the difference between the two as the residual deviation. S403. Adopt the 3σ principle to set multi-level safety assessment standards based on historical steady-state monitoring data, including safe operation threshold, early warning threshold and extreme risk threshold; S404. Calculate the comprehensive risk index by combining the adjustment range difference and residual deviation, and compare and evaluate it with the multi-level safety assessment standards.
10. The online assessment system for controllable parameters and energy consumption of thermal power units according to claim 9, characterized in that: The formula for calculating the comprehensive risk index is as follows: ； in As an ideal reference value, and For safety weighting coefficients, To adjust the difference in amplitude, This is the residual deviation; The RI value of the comprehensive risk index is negatively correlated with the safety performance of thermal power units, and is deemed unqualified when the RI value exceeds the preset limit risk threshold.