CN-121980154-A - Method and device for checking and diagnosing carbon emission multisource data fusion of coal-fired unit

Abstract

The invention relates to the technical field of energy environmental protection and carbon monitoring, and discloses a method and a device for checking and diagnosing carbon emission multisource data fusion of a coal-fired unit. Based on the comparison of the two types of deviation values and the dynamic threshold value, the abnormal model is accurately positioned, and then the multi-source data key parameters are reversely diagnosed aiming at the abnormal model. The pain point of specific abnormal parameters such as the concentration of CO 2 , the flow of flue gas, the carbon content of coal elements and the like which are difficult to lock in the prior art is effectively cracked, the transition from fuzzy judgment to accurate positioning of abnormal tracing is realized, and the data abnormal tracing efficiency and accuracy are greatly improved.

Inventors

• YANG JUN
• PA LA SHA TI
• REN XUEMEI
• LIU XIN
• GUO PAN
• WEN GUANG
• Nan Huiyun
• LI CHENGLIN
• SONG MINGGUANG
• ZHAO YUNYAN
• SUN YOUYUAN
• TIAN AIJUN
• WEI HONGXIANG
• LAN YAJUN
• YANG MINGZONG
• SHEN YUE

Assignees

• 华电新疆五彩湾北一发电有限公司
• 华电电力科学研究院有限公司

Dates

Publication Date

20260505

Application Date

20251126

Claims (10)

1. 1. The method for checking and diagnosing the carbon emission multisource data fusion of the coal-fired unit is characterized by comprising the following steps of: Respectively inputting multisource data of the coal-fired unit into a preset monitoring model, an accounting model and a prediction model, and respectively outputting a carbon emission monitoring value, a carbon emission accounting value and a carbon emission prediction value; Calculating a first deviation value between the carbon emission monitor value and the carbon emission predicted value, and a second deviation value between the carbon emission accounting value and the carbon emission predicted value; Comparing the first deviation value and the second deviation value with a preset dynamic deviation threshold value, and determining a model with data abnormality according to a comparison result; and diagnosing a plurality of key parameters of the multi-source data one by one according to the model with the data abnormality to obtain a target diagnosis report.
2. 2. The method of claim 1, wherein the inputting the multisource data of the coal-fired unit into the preset monitoring model, accounting model and prediction model, respectively, and outputting the carbon emission monitoring value, carbon emission accounting value and carbon emission prediction value, respectively, comprises: Acquiring carbon dioxide concentration, flue gas flow, flue gas humidity, furnace coal consumption, elemental carbon, industrial score data and elemental analysis data of a coal-fired unit; Generating multi-source data by combining the carbon dioxide concentration, the flue gas flow, the flue gas humidity, the consumption of the coal into the furnace, the elemental carbon, the industrial score data and the elemental analysis data; inputting the carbon dioxide concentration, the flue gas flow and the flue gas humidity of the multi-source data into a preset monitoring model, and outputting a carbon emission monitoring value; Inputting the multi-source data into a preset accounting model to output a carbon emission accounting value; And inputting the flue gas flow, the burning coal consumption, the elemental carbon, the industrial minute data and the elemental analysis data of the multi-source data into a preset prediction model, and outputting a carbon emission predicted value.
3. 3. The method of claim 1, wherein the calculating a first deviation value between the carbon emission monitor value and the carbon emission prediction value, and a second deviation value between the carbon emission accounting value and the carbon emission prediction value, comprises: calculating an absolute value of a first difference between the carbon emission monitor value and the carbon emission prediction value; Calculating the ratio between the absolute value of the first difference value and the carbon emission predicted value to generate a first deviation value; Calculating an absolute value of a second difference between the carbon emission accounting value and the carbon emission prediction value; And calculating the ratio between the absolute value of the second difference value and the carbon emission predicted value to generate a second deviation value.
4. 4. The method of claim 1, wherein comparing the first deviation value and the second deviation value with a predetermined dynamic deviation threshold value, and determining a model of the presence of data anomalies based on the comparison result comprises: When the first deviation value is larger than a preset dynamic deviation threshold value and the second deviation value is smaller than or equal to the preset dynamic deviation threshold value, determining a model with data abnormality as the monitoring model; and when the second deviation value is larger than the preset dynamic deviation threshold value and the first deviation value is smaller than or equal to the preset dynamic deviation threshold value, determining a model with data abnormality as the accounting model.
5. 5. The method according to claim 4, wherein when the model with the data anomaly is the monitoring model, the diagnosing the multiple key parameters of the multi-source data one by one according to the model with the data anomaly to obtain a target diagnosis report includes: Diagnosing the carbon dioxide concentration, the flue gas flow and the flue gas humidity of the multi-source data in sequence according to the monitoring model with the abnormal data, and generating a plurality of first diagnosis results; generating a first preliminary diagnostic report based on a plurality of the first diagnostic results; and performing secondary confirmation on the first primary diagnosis report to generate a first target diagnosis report.
6. 6. The method according to claim 4, wherein when the model with data anomalies is the accounting model, the diagnosing the multiple key parameters of the multi-source data one by one according to the model with data anomalies to obtain a target diagnosis report includes: according to the accounting model with the abnormal data, sequentially diagnosing the low-order heating value, the burning coal consumption and the elemental carbon content of the multi-source data to generate a plurality of second diagnosis results; generating a second preliminary diagnostic report based on a plurality of the second diagnostic results; And performing secondary confirmation on the second primary diagnosis report to generate a second target diagnosis report.
7. 7. The utility model provides a coal-fired unit carbon emission multisource data fuses check diagnosis device which characterized in that, the device includes: the input module is used for respectively inputting the multisource data of the coal-fired unit into a preset monitoring model, an accounting model and a prediction model and respectively outputting a carbon emission monitoring value, a carbon emission accounting value and a carbon emission prediction value; a calculation module for calculating a first deviation value between the carbon emission monitor value and the carbon emission prediction value, and a second deviation value between the carbon emission accounting value and the carbon emission prediction value; The comparison module is used for comparing the first deviation value and the second deviation value with a preset dynamic deviation threshold value, and determining a model with data abnormality according to a comparison result; And the diagnosis module is used for diagnosing a plurality of key parameters of the multi-source data one by one according to the model with the data abnormality to obtain a diagnosis result.
8. 8. An electronic device, comprising: The system comprises a memory and a processor, wherein the memory and the processor are in communication connection, the memory stores computer instructions, and the processor executes the computer instructions so as to execute the multi-source data fusion check and diagnosis method for the carbon emission of the coal-fired unit according to any one of claims 1 to 6.
9. 9. A computer-readable storage medium, wherein computer instructions for causing a computer to execute the multi-source data fusion check diagnosis method for coal-fired unit carbon emission according to any one of claims 1 to 6 are stored on the computer-readable storage medium.
10. 10. A computer program product comprising computer instructions for causing a computer to perform the multi-source data fusion check diagnosis method of coal-fired unit carbon emission of any of claims 1 to 6.

Description

Method and device for checking and diagnosing carbon emission multisource data fusion of coal-fired unit Technical Field The invention relates to the technical field of energy environmental protection and carbon monitoring, in particular to a multi-source data fusion checking and diagnosing method and device for carbon emission of a coal-fired unit. Background Global warming caused by greenhouse gas emission has become a focus of common attention of the international society, wherein CO 2 generated by thermal power generation is one of main sources of greenhouse gas, and emission control is a key link for realizing a 'two carbon' target and coping with climate change. The CO 2 emission of the coal-fired unit is accurately monitored, so that the method is not only an important basis for implementing carbon emission supervision and management by an ecological environment department, but also a core basis for optimizing emission control strategies, avoiding performance punishment risks and realizing green low-carbon transformation of thermal power enterprises. Along with the continuous improvement of the global requirement on the quality of carbon emission data, the accuracy, the instantaneity and the reliability of CO 2 emission monitoring of a coal-fired unit become core issues of industrial interest, and the construction of a scientific and effective monitoring system becomes an important subject to be solved urgently. Therefore, the traditional emission monitoring of the coal-fired unit CO 2 is a continuous monitoring method (CEMS), and the continuous monitoring method measures the flue gas parameters in real time through an automatic system to calculate the emission, but the data adopted by the method is high in instantaneity, is easily interfered by factors such as blockage of a sampling probe, drift of a sensor, failure of calibration gas, inaccurate measurement of humidity and the like, and causes difficulty in positioning specific abnormal parameters such as the concentration of CO 2, the flow of flue gas, the carbon content of coal-fired elements and the like, and difficulty in tracing the abnormal sources. Disclosure of Invention The invention provides a multi-source data fusion check diagnosis method and device for carbon emission of a coal-fired unit, which are used for solving the problems that specific abnormal parameters such as CO 2 concentration, flue gas flow, element carbon content of coal and the like are difficult to locate and the abnormal tracing is difficult in the prior art. In a first aspect, the invention provides a multi-source data fusion check diagnosis method for carbon emission of a coal-fired unit, which comprises the following steps: Respectively inputting multisource data of the coal-fired unit into a preset monitoring model, an accounting model and a prediction model, and respectively outputting a carbon emission monitoring value, a carbon emission accounting value and a carbon emission prediction value; Calculating a first deviation value between the carbon emission monitor value and the carbon emission predicted value, and a second deviation value between the carbon emission accounting value and the carbon emission predicted value; Comparing the first deviation value and the second deviation value with a preset dynamic deviation threshold value, and determining a model with data abnormality according to a comparison result; and diagnosing a plurality of key parameters of the multi-source data one by one according to the model with the data abnormality to obtain a target diagnosis report. According to the invention, multi-source data of the coal-fired unit are processed in parallel in a multi-model manner, and three types of carbon emission data including monitoring, accounting and prediction are synchronously output. Based on the comparison of the two types of deviation values and the dynamic threshold value, the abnormal model is accurately positioned, and then the multi-source data key parameters are reversely diagnosed aiming at the abnormal model. The pain point of specific abnormal parameters such as the concentration of CO 2, the flow of flue gas, the carbon content of coal elements and the like which are difficult to lock in the prior art is effectively cracked, the transition from fuzzy judgment to accurate positioning of abnormal tracing is realized, and the data abnormal tracing efficiency and accuracy are greatly improved. In an alternative embodiment, the inputting the multisource data of the coal-fired unit into the preset monitoring model, the accounting model and the prediction model respectively, and outputting the carbon emission monitoring value, the carbon emission accounting value and the carbon emission prediction value respectively includes: Acquiring carbon dioxide concentration, flue gas flow, flue gas humidity, furnace coal consumption, elemental carbon, industrial score data and elemental analysis data of a coal-fired unit; Generating multi-sourc