CN-122015570-A - SEC-based optimization algorithm for air cooling island flushing system

CN122015570ACN 122015570 ACN122015570 ACN 122015570ACN-122015570-A

Abstract

The invention discloses an optimization algorithm of an air cooling island flushing system based on SEC (SEC), which relates to the technical field of intelligent operation and maintenance of an air cooling system of a thermal power generating unit, and comprises the steps of collecting air cooling island operation state data, environment parameters and surface infrared thermal imaging images, inverting the thermal resistance distribution of dirt in each area according to the infrared thermal imaging images and the cooling air outlet temperature of a condenser, dividing flushing subareas by combining a physical structure, calculating local dirt indexes, matching an optimal flushing mode according to dirt types based on a multi-mode SEC-cleaning effect mapping library, predicting heat exchange efficiency gain and energy consumption by combining historical heat transfer performance, calculating the energy efficiency gain ratio of each subarea, setting a dynamic energy efficiency gain threshold, identifying subareas to be flushed, generating a joint control instruction containing subarea identification, mode, water pressure, flow and duration, executing differential flushing, evaluating actual gain and energy consumption after flushing, and carrying out on-line correction on the multi-mode SEC cleaning effect mapping library and the dynamic energy efficiency gain threshold, thereby realizing closed loop self-optimization.

Inventors

WEI WEI
LIANG GUOQIANG
ZHANG HAIHONG
QIAO WENYI
HONG DAZHI
LIN XIANCHAO

Assignees

内蒙古大唐国际托克托发电有限责任公司

Dates

Publication Date: 20260512
Application Date: 20251215

Claims (10)

1. An optimization algorithm of an air cooling island flushing system based on SEC is characterized by comprising the following steps: Acquiring air cooling island running state data, environmental parameters and surface infrared thermal imaging images; inverting the dirt thermal resistance distribution of each area of the air cooling island according to the infrared thermal imaging image and the cooling air outlet temperature of the condenser, dividing a plurality of flushing subareas by combining the physical structure of the air cooling island, and calculating the local dirt index of each flushing subarea; Determining dirt types according to local dirt indexes of each flushing subarea based on a pre-constructed multi-mode SEC cleaning effect mapping library, and matching corresponding optimal flushing modes for each flushing subarea; according to the historical heat transfer performance data, predicting heat exchange efficiency gains of all flushing subareas after adopting subareas to correspond to an optimal flushing mode, and calculating energy efficiency gain ratio of all flushing subareas by combining with an energy consumption estimated value in a flushing process; setting a dynamic energy efficiency gain threshold, comparing the energy efficiency gain ratio of each flushing subarea with the dynamic energy efficiency gain threshold, and identifying the flushing subarea meeting the flushing triggering condition; Aiming at the identified flushing subarea, generating a combined flushing control instruction comprising a subarea mark, an optimal flushing mode, a target water pressure, a target water flow and a flushing duration time, and issuing the combined flushing control instruction to a flushing executing mechanism to execute differential flushing; After differential flushing is completed, based on the condenser cooling air outlet temperature and the infrared thermal imaging image acquired after flushing, the actual heat exchange efficiency gain and the actual flushing energy consumption are estimated, and the multi-mode SEC cleaning effect mapping library and the dynamic energy efficiency gain threshold are corrected on line.
2. The optimization algorithm of the SEC-based air cooling island flushing system as set forth in claim 1, wherein the steps of collecting air cooling island operation state data, environment parameters and surface infrared thermal imaging images are as follows: acquiring a current steam saturation temperature through a temperature sensor arranged on a condenser system The steam saturation temperature is determined by a unit backpressure meter; obtaining the actually measured outlet temperature of cooling air through a temperature probe arranged in an outlet air duct of an air cooler The electric energy monitoring module is used for collecting an input electric power signal of the flushing water pump in real time; The method comprises the steps of obtaining instantaneous volume flow of flushing water through an electromagnetic flowmeter, and synchronously calling a factory-level weather station interface to obtain current ambient wind speed, relative humidity and atmospheric dust concentration; and starting an infrared thermal imaging array erected above the air cooling island platform, scanning the surfaces of all air cooling fins in a half-minute period, and outputting a two-dimensional surface temperature distribution image subjected to radiation calibration and non-uniformity correction.
3. The optimization algorithm of the SEC-based air cooling island flushing system according to claim 2, wherein the method is characterized by inverting the dirt thermal resistance distribution of each area of the air cooling island according to the infrared thermal imaging image and the cooling air outlet temperature of the condenser, dividing a plurality of flushing subareas by combining the physical structure of the air cooling island, and calculating the local dirt index of each flushing subarea, and comprises the following specific steps: Dividing the whole air cooling island into a plurality of non-overlapping geometric flushing subareas according to the air cooling island fan unit layout and the tube bundle arrangement mode, wherein each subarea corresponds to one subarea in the infrared image; Calculating the average surface temperature of each sub-area; based on a steady-state heat transfer principle, inverting equivalent dirt thermal resistance of a partition by utilizing steam saturation temperature, cooling air outlet temperature, average surface temperature and known clean state convection heat transfer coefficient of equipment; The current ambient wind speed and dust concentration are combined, the dirt thermal resistance is subjected to environment weighted correction, and the local dirt index of the partition is obtained, wherein the formula is as follows: ; Wherein, the For the thermal resistance of the dirt, The temperature of the saturated steam is set to be the saturation temperature of the steam, Is the first The average surface temperature of the zones is determined, In order to cool the air outlet temperature, The standard convective heat transfer coefficient is the inherent parameter of the equipment in the clean state of the air cooling fin.
4. The optimization algorithm of the SEC-based air cooling island flushing system according to claim 3, wherein the method is characterized in that the method comprises the steps of determining the type of dirt according to the local dirt index of each flushing subarea based on a pre-constructed multi-mode SEC-cleaning effect mapping library, and matching the corresponding optimal flushing mode for each flushing subarea, and comprises the following specific steps: Before the system is put into operation, cleaning effects of three flushing modes including high-pressure jet flow, low-pressure spraying and ultrasonic atomization are respectively tested for typical dirt types through field experiments, flushing energy consumption and end difference improvement quantity under each mode are recorded, and unit benefit energy consumption ratio is calculated; Organizing the unit benefit energy consumption ratio into a multi-mode SEC cleaning effect mapping library according to the dirt type and the flushing mode; in the operation stage, inputting the local dirt index of the current partition into a pre-trained dirt classification model, and outputting a corresponding dirt class; and querying a mapping library, obtaining the unit benefit energy consumption ratio of the dirt category in each flushing mode, and selecting the flushing mode with the minimum unit benefit energy consumption ratio as the optimal flushing mode of the partition.
5. The optimization algorithm of the SEC-based air cooling island flushing system of claim 4, wherein the heat exchange efficiency gain of each flushing partition after adopting the partition corresponding to the optimal flushing mode is predicted according to the historical heat transfer performance data, and the energy efficiency gain ratio of each flushing partition is calculated by combining the flushing process energy consumption estimated value, and the specific steps are as follows: establishing a prediction model between the thermal resistance of dirt and the improvement amount of end difference based on a historical operation database of the power plant; the dirt thermal resistance and the optimal flushing mode of the current subarea are used as inputs, the improvement amplitude of the end difference after cleaning is predicted, and the predicted end difference improvement quantity is converted into the heat exchange efficiency gain of the condenser, wherein the heat exchange efficiency gain End difference improvement amount The conversion relation of (2) is: ; Wherein, the Is used as a reference value of the heat efficiency of the current condenser, The current condenser heat efficiency reference value and the actual end difference are acquired in real time for the current actual end difference And (3) with Calculating to obtain; Estimating the total energy consumption required by the current flushing according to the energy consumption coefficient of the unit flow corresponding to the optimal flushing mode, the target water flow and the recommended flushing time, wherein the total energy consumption required by the flushing is the washing energy consumption The estimation formula of (2) is: ; Wherein, the To and optimum flushing mode The corresponding energy consumption coefficient of the unit flow, For a target water flow rate, A recommended flushing period; Dividing the heat exchange efficiency gain by the estimated energy consumption to obtain a partitioned energy efficiency gain ratio, wherein the energy efficiency gain ratio characterizes the heat efficiency improvement brought by unit flushing energy consumption, and judging whether the flushing is worth according to the energy efficiency gain ratio, wherein the energy efficiency gain ratio is equal to the energy efficiency gain ratio The formula is: ; Wherein, the In order to gain the heat exchange efficiency, Total energy consumption required for flushing.
6. The optimization algorithm of the SEC-based air cooling island flushing system according to claim 5, wherein the setting of the dynamic energy efficiency gain threshold value, comparing the energy efficiency gain ratio of each flushing subarea with a reference convection heat transfer coefficient, and identifying the flushing subarea meeting the flushing trigger condition comprises the following specific steps: Initializing a dynamic energy efficiency gain threshold value to be a sliding average value of energy efficiency gain ratios in historical effective flushing events, and updating the dynamic energy efficiency gain threshold value by adopting an exponential smoothing method according to the actual energy efficiency gain ratio of the last several effective flushing in each flushing decision period The update rule of (2) is: ; Wherein, the Is the cause of forgetfulness, For the energy efficiency-to-benefit ratio average of the most recent active flush event, Is shown in the first When the flushing decision is executed for the second time, the energy efficiency gain judgment standard adopted by the system is adopted; And comparing the energy efficiency and gain ratio calculated by each partition with a current dynamic threshold, if the energy efficiency and gain ratio of a certain partition is higher than the current dynamic threshold, judging that the current flushing has forward net energy efficiency and gain, and incorporating the partition into the partition set to be flushed.
7. The optimization algorithm of the SEC-based air cooling island flushing system of claim 6, wherein the generating of the combined flushing control command including the partition identification, the optimal flushing mode, the target water pressure, the target water flow and the flushing duration for the identified flushing partition and issuing to the flushing execution mechanism to execute the differential flushing comprises the following specific steps: Traversing each partition in the partition set to be washed, and extracting technological parameters related to the optimal washing mode from the multi-mode SEC cleaning effect mapping library, wherein the technological parameters comprise target washing water pressure, target washing water flow and recommended washing duration; packaging the partition physical identifier, the flushing mode number, the target water pressure, the target water flow and the flushing time length into a structural control instruction; All control instructions are issued to local execution units of corresponding partitions in parallel through an industrial communication protocol, and the local execution units comprise an electric regulating valve, a variable-frequency water pump driver and a nozzle start-stop controller; The execution unit independently adjusts the opening degree of the valve, the rotating speed of the water pump and the opening and closing state of the nozzle according to the instruction.
8. The optimization algorithm of the SEC-based air cooling island flushing system of claim 7, wherein after differential flushing is completed, based on the condenser cooling air outlet temperature and the infrared thermal imaging image acquired after flushing, the actual heat exchange efficiency gain and the actual flushing energy consumption are estimated, and accordingly, the multi-mode SEC cleaning effect mapping library and the dynamic energy efficiency gain threshold are corrected on line, and the specific steps are as follows: after all zone flushing operations are completed and the thermal force field is waited to be stable, the outlet temperature of the cooling air is collected again Infrared images on the surfaces of the air cooling islands; calculating the actual end difference improvement quantity , wherein, In order to cool the air outlet temperature before flushing, Cooling the air outlet temperature after flushing; and deducing the actual heat exchange efficiency gain according to the actual end difference improvement quantity The formula is: ; Wherein, the Is used as a reference value of the heat efficiency of the current condenser, For the actual end difference before flushing, The actual end difference improvement amount; Synchronously reading an integral value of the electric power of the water pump in the flushing process to obtain the actual flushing energy consumption And calculates the actual energy efficiency gain ratio The formula is: ; Wherein, the The electric energy actually consumed in the flushing execution stage is obtained by integrating the electric energy monitoring module in real time, The actual heat exchange efficiency is increased; If the relative deviation between the actual energy efficiency gain ratio and the predicted value exceeds a preset tolerance, triggering a model correction mechanism, wherein the model correction mechanism is used for updating the unit benefit energy consumption ratio of the corresponding dirt type and the flushing mode in the mapping library or taking the actual energy efficiency gain ratio of the effective flushing into a statistical sample of a dynamic threshold.
9. A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the computer program is characterized in that the processor realizes the steps of the SEC-based air cooling island flushing system optimization algorithm in any one of claims 1-8 when executing the computer program.
10. A computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the steps of the SEC-based air cooling island flushing system optimization algorithm of any one of claims 1-8.

Description

SEC-based optimization algorithm for air cooling island flushing system Technical Field The invention relates to the technical field of intelligent operation and maintenance of an air cooling system of a thermal power generating unit, in particular to an optimization algorithm of an air cooling island flushing system based on SEC. Background The air cooling island is used as core heat exchange equipment of the direct air cooling unit, and the operation performance of the air cooling island directly influences the vacuum degree of the condenser and the coal consumption level of the unit. With the great construction of air-cooled thermal power generating units in western coal-rich and water-deficient areas of China, the problem of dirt deposition of air-cooled islands in high-dust and strong-wind sand environments is increasingly prominent, so that the heat exchange efficiency is reduced, the backpressure is increased, and the power supply coal consumption is increased. To maintain the performance of the air cooling system, regular flushing has become a conventional operation and maintenance means. In the existing flushing strategy, timing control or start-stop logic based on a fixed end difference threshold value is mostly adopted, and part of advanced systems introduce infrared temperature measurement to assist in judging the overall pollution degree, but flushing decision is still based on whether dirtying is the only basis, and quantitative evaluation of energy efficiency cost of flushing behavior per se is lacking. In recent years, although research is attempted to take water consumption or pumping work as an optimization target, a systematic intelligent control framework with unit energy consumption benefits as a core has not been formed. The prior art system still has defects, such as that a flushing trigger mechanism does not consider the balance of flushing income and energy consumption cost, so that low-efficiency and even negative-income flushing is often caused, electric energy and water resources are wasted, accurate cleaning according to needs cannot be realized, a flushing strategy lacks closed loop feedback and self-adaptive capacity, the flushing strategy is difficult to adjust along with equipment aging, seasonal variation or water quality fluctuation, and the attenuation of an optimization effect is obvious under long-term operation. Disclosure of Invention The present invention has been made in view of the above-described problems occurring in the prior art. Therefore, the invention provides an optimization algorithm of an air cooling island flushing system based on SEC, which solves the problems that a flushing trigger mechanism does not consider the balance of flushing income and energy consumption cost, so that low-efficiency and even negative-income flushing is often caused, electric energy and water resources are wasted, accurate cleaning as required cannot be realized, a flushing strategy lacks closed loop feedback and self-adaptation capability, the flushing strategy is difficult to adjust along with equipment aging, seasonal variation or water quality fluctuation, and the attenuation of an optimization effect is obvious under long-term operation. In order to solve the technical problems, the invention provides the following technical scheme: In a first aspect, the present invention provides an air cooling island flushing system optimization algorithm based on SEC, including: Acquiring air cooling island running state data, environmental parameters and surface infrared thermal imaging images; inverting the dirt thermal resistance distribution of each area of the air cooling island according to the infrared thermal imaging image and the cooling air outlet temperature of the condenser, dividing a plurality of flushing subareas by combining the physical structure of the air cooling island, and calculating the local dirt index of each flushing subarea; Determining dirt types according to local dirt indexes of each flushing subarea based on a pre-constructed multi-mode SEC cleaning effect mapping library, and matching corresponding optimal flushing modes for each flushing subarea; according to the historical heat transfer performance data, predicting heat exchange efficiency gains of all flushing subareas after adopting subareas to correspond to an optimal flushing mode, and calculating energy efficiency gain ratio of all flushing subareas by combining with an energy consumption estimated value in a flushing process; setting a dynamic energy efficiency gain threshold, comparing the energy efficiency gain ratio of each flushing subarea with the dynamic energy efficiency gain threshold, and identifying the flushing subarea meeting the flushing triggering condition; Aiming at the identified flushing subarea, generating a combined flushing control instruction comprising a subarea mark, an optimal flushing mode, a target water pressure, a target water flow and a flushing duration time, and is