CN-122022570-A - Actual measurement energy consumption management system of coal-fired power unit

CN122022570ACN 122022570 ACN122022570 ACN 122022570ACN-122022570-A

Abstract

The invention discloses an actually measured energy consumption management system of a coal-fired power unit, and particularly relates to the field of energy consumption management, comprising an acquisition module, a preprocessing module, a characteristic module, an analysis module, a deviation module, a strategy module, an optimization module and an early warning module; the system comprises an acquisition module, a preprocessing module, an analysis module and a control module, wherein the acquisition module acquires multi-dimensional parameters of unit operation, unifies parameter acquisition frequencies, the preprocessing module identifies abnormal data based on a3 sigma principle, repairs the abnormal data by adopting linear interpolation, performs standardized processing on repaired data, the analysis module analyzes the operation efficiency of core equipment, acquires basic energy consumption based on the equipment efficiency and the operation parameters, and obtains total actual energy consumption by additional energy consumption correction.

Inventors

Mou Qingsong
WU MINGZONG
Gong Mengjuan
GUO YANLONG
CAO DA
REN XUEJUN
LI ZHEN
Lawula Mai'erhan
WANG DONG
JIA LIN

Assignees

华电新疆准东五彩湾发电有限公司

Dates

Publication Date: 20260512
Application Date: 20260123

Claims (9)

1. An actual measurement energy consumption management system of a coal-fired power unit, which is characterized by comprising: The acquisition module is used for acquiring the multi-dimensional parameters of the unit operation and unifying the parameter acquisition frequency; The preprocessing module is used for identifying abnormal data based on a3 sigma principle, repairing the abnormal data by adopting linear interpolation and carrying out standardized processing on the repaired data; the feature module is used for constructing a multidimensional working condition feature vector, performing cosine similarity matching with a standard working condition library and generating the feature vector by adopting weighted fusion on the transition working condition; the analysis module is used for analyzing the operation efficiency of the core equipment, acquiring basic energy consumption based on the equipment efficiency and the operation parameters, and obtaining total actual measured energy consumption through additional energy consumption correction; the deviation module is used for analyzing the energy consumption deviation quantity, quantifying the contribution degree of each operation parameter to the energy consumption deviation by adopting a partial derivative method and determining the main cause of the energy consumption deviation; the strategy module is used for analyzing based on the energy consumption deviation and main causes, and making a differential operation parameter strategy to give consideration to the safety of unit equipment; the optimization module is used for evaluating the optimization effect coefficient, judging whether to update the optimal operation related parameter library based on the coefficient, and correcting the coefficient in the parameter strategy again if the optimization requirement is not met; and the early warning module is used for training an LSTM neural network energy consumption prediction model, predicting the energy consumption value of the future period and sending early warning signals and preliminary optimization suggestions based on prediction deviation.
2. The measured energy consumption management system of the coal-fired power unit according to claim 1, wherein the unit operation multidimensional parameter comprises a boiler outlet steam temperature Pressure of main steam Flow rate of water supply pump Generating load of unit The coal receives the basic low-position heating value Ash content of coal Temperature of exhaust gas Air supply quantity The unified parameter acquisition frequency is set to be 1 time/min, so that timeliness and synchronism of all parameter acquisition are ensured, and collaborative data support is provided for follow-up accounting, identification and optimization links.
3. The system for managing actual measurement energy consumption of coal-fired power unit according to claim 2, wherein the 3 sigma principle recognition of abnormal data is to calculate the average value of each parameter acquisition sequence And standard deviation If a certain time data point Satisfy the following requirements Judging the data point as abnormal data, and obtaining repaired data by the linear interpolation repairing abnormal data Wherein Is valid data at a time before the abnormal data, Effective data at the moment after abnormal data is obtained, wherein the formula of the normalized processing of the repaired data is as follows Interference of parameter dimension difference on subsequent calculation is eliminated through the processing, and a standardized unit operation multidimensional parameter data sequence is obtained 。
4. The actually measured energy consumption management system of the coal-fired power unit according to claim 3, wherein the multidimensional working condition feature vector is as follows The core characteristics of the load, the coal quality and the running state of the equipment are integrated, and the standard working condition library contains low-load working conditions Working condition of medium load High load condition Obtaining a similarity value after the cosine similarity is matched with a standard working condition library Wherein Is the k standard working condition vector in the standard working condition library, Representing a vector dot product of the vector, The vector module length is represented, if the maximum similarity value is more than or equal to 0.85, the corresponding standard working condition is judged, and the formula for generating the feature vector by weighting and fusing the transition working conditions is as follows Wherein 、 And the full-coverage recognition of the working condition is realized for the similarity between the current vector and the two standard working condition vectors with the highest similarity.
5. The actually measured energy consumption management system of the coal-fired power unit according to claim 2, wherein the operating efficiency of the core equipment comprises boiler efficiency Efficiency of steam turbine Efficiency of generator The basic energy consumption The core energy consumption quantification is realized based on the power generation load and the equipment efficiency, and the additional energy consumption correction total actual measured energy consumption comprises smoke exhaust loss energy consumption And air supply loss energy consumption Total measured energy consumption 。
6. The actually measured energy consumption management system of the coal-fired power unit according to claim 5, wherein the energy consumption deviation amount is as follows Wherein The optimal energy consumption threshold value generated based on the historical optimal operation data fitting under the current working condition is obtained by the partial derivative method quantization parameter contribution degree formula Wherein The quantized contribution degree of the ith operation parameter to the measured energy consumption deviation of the unit is obtained, For the ith parameter in the multi-dimensional parameters of the unit operation, For the partial derivative of the energy consumption deviation with respect to the parameter, The main cause of the energy consumption deviation is to sort the contribution degrees of all the parameters according to absolute values, and the parameter with the largest absolute value of the contribution degrees is taken as the main cause.
7. The system for managing actual energy consumption of a coal-fired power unit as set forth in claim 6, wherein the analysis of the energy consumption deviation and the main cause is to set a deviation threshold Judging And (3) with The strategy direction is formulated by combining the main causes, and the differential operation parameter strategy comprises the following steps of when Maintaining current parameters and updating energy consumption accounting results every 5 minutes when In the case of the main cause being If the temperature is too high, the rotating speed of the induced draft fan is changed into If the main cause is If the pressure is too large, changing the opening of the baffle of the blower to be If the main cause is If the ratio of the fire coal is too low, the ratio of the fire coal is changed into The compromise of equipment safety refers to when When the water supply pump flow is changed to And equipment damage caused by low energy consumption is avoided.
8. The actually measured energy consumption management system of the coal-fired power unit according to claim 5, wherein the optimization effect coefficient is as follows Wherein The updating and judging of the optimal operation related parameter library means to set an optimal qualification threshold value If (if) And judging that the optimization is effective, storing the current working condition parameters, the strategy and the energy consumption data into an optimal operation database, and synchronously updating the standard working condition feature library and the optimal energy consumption threshold library.
9. The system for managing actual measurement energy consumption of a coal-fired power unit as set forth in claim 7, wherein the LSTM neural network energy consumption prediction model training is based on actual measurement energy consumption data, working condition data and optimization records of 6 months history, and is performed at 24 past moments 、、、、、 Model training is carried out for the input layer, wherein the future period energy consumption value prediction refers to outputting the energy consumption predicted values of 12 future moments through the trained model The early warning signal and the preliminary optimization suggestion pushing means calculating a prediction deviation If (if) The control center immediately sends out a high-energy consumption early warning signal and synchronously pushes the preliminary optimization suggestion, so that prospective risk management and control are realized.

Description

Actual measurement energy consumption management system of coal-fired power unit Technical Field The invention relates to the technical field of energy consumption management, in particular to an actual measurement energy consumption management system of a coal-fired power unit. Background In the coal-fired power generation, energy consumption management is a core link for guaranteeing economic operation of a unit and reducing carbon emission, technical means such as manual recording of key operation parameters, single sensor monitoring of energy consumption indexes, basic data statistical analysis and the like are generally adopted in the current industry, preliminary management and control of energy consumption of the coal-fired power unit are realized, and related technologies are widely applied to the scenes such as energy consumption data acquisition, statistical accounting and preliminary cost estimation. The existing energy consumption management technology can meet basic energy consumption data recording and simple accounting requirements, provides data reference for unit operation to a certain extent, but has the obvious defects that firstly, the data acquisition dimension is single, the data acquisition dimension is focused on core indexes such as generated energy and coal consumption, the auxiliary parameters such as boiler outlet steam temperature, exhaust gas temperature and air supply quantity are lacked to cooperatively acquire, the input variables of an energy consumption accounting model are incomplete, the deviation of an accounting result is large, and the decision requirement of fine energy consumption management and control is difficult to support. In addition, the prior art lacks a full-period energy consumption trend prediction and early warning mechanism, cannot avoid high-energy consumption operation risks in advance, and restricts the realization of energy consumption cost optimization and sustainable operation targets. Aiming at the defects of the prior art in the aspects of data acquisition integrity, accounting accuracy, working condition adaptability, optimization intellectualization and the like, the scheme provides an actual measurement energy consumption management system of a coal-fired power unit, which solves the core problems of the prior art by combining data analysis, decision support and full-flow management and control concepts through multi-dimensional parameter acquisition, refined data processing, accurate working condition identification, dynamic deviation analysis and closed-loop optimization. Disclosure of Invention In order to overcome the defects in the prior art, the embodiment of the invention provides an actual measurement energy consumption management system of a coal-fired power unit, which solves the problems in the background art through the following scheme. In order to achieve the purpose, the invention provides the following technical scheme that the actually measured energy consumption management system of the coal-fired power unit comprises: The acquisition module is used for acquiring the multi-dimensional parameters of the unit operation and unifying the parameter acquisition frequency; The preprocessing module is used for identifying abnormal data based on a3 sigma principle, repairing the abnormal data by adopting linear interpolation and carrying out standardized processing on the repaired data; the feature module is used for constructing a multidimensional working condition feature vector, performing cosine similarity matching with a standard working condition library and generating the feature vector by adopting weighted fusion on the transition working condition; the analysis module is used for analyzing the operation efficiency of the core equipment, acquiring basic energy consumption based on the equipment efficiency and the operation parameters, and obtaining total actual measured energy consumption through additional energy consumption correction; the deviation module is used for analyzing the energy consumption deviation quantity, quantifying the contribution degree of each operation parameter to the energy consumption deviation by adopting a partial derivative method and determining the main cause of the energy consumption deviation; the strategy module is used for analyzing based on the energy consumption deviation and main causes, and making a differential operation parameter strategy to give consideration to the safety of unit equipment; the optimization module is used for evaluating the optimization effect coefficient, judging whether to update the optimal operation related parameter library based on the coefficient, and correcting the coefficient in the parameter strategy again if the optimization requirement is not met; and the early warning module is used for training an LSTM neural network energy consumption prediction model, predicting the energy consumption value of the future period and sending early warning signals and pr