CN113094988A - Data-driven slurry circulating pump operation optimization method and system - Google Patents
Data-driven slurry circulating pump operation optimization method and system Download PDFInfo
- Publication number
- CN113094988A CN113094988A CN202110368339.7A CN202110368339A CN113094988A CN 113094988 A CN113094988 A CN 113094988A CN 202110368339 A CN202110368339 A CN 202110368339A CN 113094988 A CN113094988 A CN 113094988A
- Authority
- CN
- China
- Prior art keywords
- data
- circulating pump
- slurry circulating
- slurry
- historical
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
Abstract
The invention belongs to the technical field of desulfurization systems, and provides a method and a system for optimizing operation of a slurry circulating pump based on data driving. The method comprises the following steps: acquiring historical operation data of a slurry circulating pump; clustering the unit operation conditions based on historical operation data of the slurry circulating pump by adopting an FCM algorithm, and performing cross combination on clustering results to obtain the condition data of different operation combinations; automatically acquiring historical optimal operation data in the working condition data by adopting a comprehensive evaluation function; establishing a slurry circulating pump operation optimization model by adopting an SVM classifier based on historical optimal operation data; inputting the real-time operation data of the slurry circulating pump into the slurry circulating pump operation optimization model, and outputting the slurry circulating pump operation suggestion combination scheme.
Description
Technical Field
The invention belongs to the technical field of thermal power plant desulfurization systems, and particularly relates to a method and a system for optimizing operation of a slurry circulating pump based on data driving.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
The slurry circulating pump is a main power consumption device in a limestone-gypsum wet flue gas desulfurization system of a thermal power plant, and the realization of the operation optimization of the slurry circulating pump has important practical significance for reducing the power consumption of the power plant and improving the economic benefit of the power plant.
At present, the operation mode of a slurry circulating pump in a domestic desulfurization system of a thermal power plant is mainly controlled by an operator manually. In order to ensure that the concentration discharge of the sulfur dioxide at the outlet meets the environmental protection requirement, operators generally increase the number of the slurry circulating pumps to provide sufficient amount of limestone slurry and even excessive amount of limestone slurry for the absorption reaction of the sulfur dioxide in the absorption tower, and the power consumption in the desulfurization process is increased in a step shape when one slurry circulating pump is added, so that the slurry circulating pump has huge energy-saving potential.
The energy-saving optimization research on the desulfurization system slurry circulating pump mainly takes a reaction mechanism model as a main part, but the chemical reaction in the absorption tower is complex and dynamic balance, has the characteristics of time-varying property, large inertia, hysteresis, nonlinearity and the like, and the mechanism modeling is difficult to reflect the complex characteristic. With the wide establishment of a plant-level monitoring information platform of a power plant in recent years, a large amount of historical operating data has been accumulated, and a data basis is provided for the operation optimization of a data-driven system. The method for establishing the working condition library of the power plant desulfurization system by combining an information Entropy theory with a K-mean and Fuzzy C-mean EKFCM (K-Means and Fuzzy C-Means combined with control the algorithm) is provided by the Hui et al; the invention provides a unit optimization operation optimizing method based on working condition division, and the method uses uncontrollable boundary condition unit load and environment temperature as objects to realize unit working condition division by fuzzy clustering method; liuyanquan et al propose a 330MW unit wet flue gas desulfurization control system target value optimization, utilize the correlation characteristic among the operating parameters of the desulfurization system, introduce the competition agglomeration algorithm into dividing the boundary and carry on the mining of the association rule; wangshan et al propose thermal power plant thermoelectric load distribution optimization by using a particle swarm algorithm, and establish a unit energy consumption analysis model and a power plant load optimization distribution model based on the particle swarm algorithm. However, the literature for researching the operation optimization of the slurry circulating pump by adopting the data mining technology is less at present.
Disclosure of Invention
Aiming at the problem of energy consumption in the current operating situation of the slurry circulating pump of the thermal power plant, the invention provides a method and a system for optimizing the operation of the slurry circulating pump based on data driving, wherein working condition division is respectively carried out by using an FCM algorithm according to unit load and inlet sulfur dioxide concentration, and then a plurality of working condition clusters are obtained by adopting a cross-combination working condition division method; based on main factors influencing the operation mode of the slurry circulating pump, automatically searching optimal historical operation data in each working condition by adopting a combined evaluation method of hierarchical analysis and entropy weight evaluation weighted fusion; establishing a slurry circulating pump operation optimization model by using the acquired historical optimal working condition data through a Support Vector Machine (SVM) algorithm, and simultaneously performing parameter optimization on the optimization model through a genetic algorithm; and inputting real-time operation data of the desulfurization system into the established operation optimization model of the slurry circulating pump, and providing an operation suggestion of the slurry circulating pump according to the operation result of the model so as to realize the operation optimization of the slurry circulating pump.
In order to achieve the purpose, the invention adopts the following technical scheme:
the first aspect of the invention provides a method for optimizing the operation of a slurry circulating pump based on data driving.
A method for optimizing operation of a slurry circulating pump based on data driving comprises the following steps:
acquiring historical operation data of a slurry circulating pump;
clustering the unit operation conditions based on historical operation data of the slurry circulating pump by adopting an FCM algorithm, and performing cross combination on clustering results to obtain the condition data of different operation combinations;
automatically acquiring historical optimal operation data in the working condition data by adopting a comprehensive evaluation function;
establishing a slurry circulating pump operation optimization model by adopting an SVM classifier based on historical optimal operation data;
inputting the real-time operation data of the slurry circulating pump into the slurry circulating pump operation optimization model, and outputting the slurry circulating pump operation suggestion combination scheme.
A second aspect of the invention provides a data-driven slurry circulation pump operation optimization system.
A data-driven slurry circulation pump operation optimization system, comprising:
a data acquisition module configured to: acquiring historical operation data of a slurry circulating pump;
a clustering module configured to: clustering the unit operation conditions based on historical operation data of the slurry circulating pump by adopting an FCM algorithm, and performing cross combination on clustering results to obtain the condition data of different operation combinations;
a comprehensive evaluation module configured to: automatically acquiring historical optimal operation data in the working condition data by adopting a comprehensive evaluation function;
a model building module configured to: establishing a slurry circulating pump operation optimization model by adopting an SVM classifier based on historical optimal operation data;
an output module configured to: inputting the real-time operation data of the slurry circulating pump into the slurry circulating pump operation optimization model, and outputting the slurry circulating pump operation suggestion combination scheme.
A third aspect of the invention provides a computer-readable storage medium.
A computer-readable storage medium, having stored thereon a computer program which, when being executed by a processor, carries out the steps of the data-driven based slurry circulation pump operation optimization method according to the first aspect.
A fourth aspect of the invention provides a computer apparatus.
A computer apparatus comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor when executing the program implementing the steps in the data-driven based slurry circulation pump operation optimization method according to the first aspect.
Compared with the prior art, the invention has the beneficial effects that:
the invention adopts the operation optimization model of the desulfurization slurry circulating pump inputted with real-time operation data, and realizes the operation suggestion of the slurry circulating pump according to the operation result of the model, thereby realizing the operation optimization.
The invention adopts FCM algorithm and a combined evaluation method based on hierarchical analysis and entropy weight weighted fusion to automatically optimize the historical operation of the slurry circulating pump under different working conditions; training an SVM classifier by using the optimal historical operation record, and performing parameter optimization by using a genetic algorithm; and inputting the established operation optimization model of the slurry circulating pump by real-time operation data, and providing an operation suggestion of the slurry circulating pump according to the operation result of the model, thereby realizing the operation optimization of the slurry circulating pump.
Simulation experiments carried out by measured data of a certain power plant show that: by using the operation optimization method of the slurry circulating pump, the power consumption of the slurry circulating pump can be reduced by about 21.55% to the maximum extent, and the method has a reference significance for energy conservation and emission reduction work of a thermal power plant.
Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
FIG. 1 is a flow chart of a method for optimizing operation of a data-driven slurry circulation pump according to an embodiment of the present invention;
FIG. 2 is a pie chart of unit load condition division in the embodiment of the invention;
FIG. 3 is a graphical illustration of a pie chart depicting inlet sulfur dioxide concentration conditions in an embodiment of the present invention;
FIG. 4 is a diagram of a hierarchy model in an embodiment of the present invention;
FIG. 5 is a graph of genetic algorithm fitness in an embodiment of the present invention;
FIG. 6 is a comparison graph of energy saving and consumption reduction after optimization in the embodiment of the present invention.
Detailed Description
The invention is further described with reference to the following figures and examples.
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise; furthermore, it will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof.
Example one
As shown in fig. 1, this embodiment provides a method for optimizing the operation of a slurry circulation pump based on data driving, and in this embodiment, the method includes the following steps:
s101: acquiring historical operation data of a slurry circulating pump;
after obtaining the historical operation data of the slurry circulating pump, the method comprises the following steps: preprocessing historical operating data and rejecting abnormal data; after data preprocessing, acquiring key influence factors by using a quantitative and qualitative combined mode, and firstly, preliminarily determining the characteristics influencing the operating efficiency of the slurry circulating pump by combining desulfurization process mechanism analysis and production experience; through mutual information and correlation coefficient calculation, main factors influencing the operation of the slurry circulating pump are quantitatively determined; the historical operation data of the slurry circulating pump comprises unit load operation data, inlet sulfur dioxide operation data, absorption tower inlet flue gas converted flow, absorption tower inlet pressure, absorption tower outlet flue gas converted flow, absorption tower outlet pressure and discharged flue gas flow rate.
S102: clustering the unit operation conditions based on historical operation data of the slurry circulating pump by adopting an FCM algorithm, and performing cross combination on clustering results to obtain the condition data of different operation combinations;
specifically, the FCM algorithm is the most widely and successfully applied fuzzy clustering algorithm at present, and the objective of clustering data is achieved by optimizing the objective function to obtain the membership of each data point to all class centers. Assume that the sample set is X ═ X1,x2,…,xnD, the classification number is c, and the membership matrix U is UijI ═ 1,2, …, c; j is 1,2, …, n, wherein uijIs the membership value of the j-th data belonging to the class i. FCM objective function of
The constraint condition is
In the formula: m is a fuzzy weighting factor greater than 1, typically 2. v. ofiI 1, 2.. c denotes the center of the ith cluster, | | xj-vi||2Representing the euclidean distance of data point j to the center of the ith cluster.
The Lagrange multiplier method is used for constructing a new objective function of
Obtaining an iterative formula of the clustering center and the membership degree by solving the extreme value of the formula (2) until u(k+1)-uk<Beta (a set threshold), stopping iteration and finishing the clustering optimization process.
And evaluating the working condition division effect by adopting the Xie-Beni index, and determining the number of clustering clusters (working condition number).
The operating condition division effect of the FCM algorithm is evaluated by using the Xie-Beni index, the lower the Xie-Beni index is, the better the clustering effect is, and the calculation formula is
In the formula: c is the number of clusters; n represents the number of data, uijIs the degree of membership, v, of data points j in class ii-xjIs a data point xjTo the center of the cluster viThe distance of (c).
S103: automatically acquiring historical optimal operation data in the working condition data by adopting a comprehensive evaluation function;
specifically, in order to automatically obtain the optimal historical operating data, evaluation functions need to be designed for the clustered historical data (different working conditions). The current common evaluation methods comprise a subjective weighting evaluation method and an objective weighting evaluation method. The subjective weighting method reflects that the appraisers pay attention to different indexes and is easily influenced by subjective factors. The objective weighting method determines the weight according to the correlation among the factors independently of the subjective preference of people, but the result is easy to deviate from the actual situation by depending too much on data. The embodiment provides a comprehensive evaluation method combining subjectivity and objectivity, and the comprehensive evaluation method based on an analytic hierarchy process and an entropy weight method is used for evaluating the running state of the slurry circulating pump, so that the advantages of the subjective and objective evaluation method are taken into consideration, and the accuracy of an evaluation result is improved.
As one or more embodiments, the process of constructing the composite evaluation function includes:
based on an analytic hierarchy process and an entropy weight method, a multiplier synthesis method is adopted to construct a comprehensive evaluation function.
S1031: analytic hierarchy process
The hierarchical analysis theory is a weight determination method widely applied and is suitable for complex comprehensive evaluation systems. When the analytic hierarchy process is used for building the system model, the method generally comprises the following steps.
(1) And constructing a hierarchical structure model.
And (3) firstly constructing a hierarchical structure model according to the influence factors by using an analytic hierarchy process, and dividing the target, the influence factors and the optimized object into a target layer, a criterion layer and a scheme layer.
(2) And constructing a judgment matrix.
Building hierarchical model objectsThe judgment matrix is the weight of each influence factor relative to the target factor according to the mutual relation among the influence factors, and the recognition of a decision maker on the importance of each influence factor is reflected. Comparing the different influencing factors to obtain aijFilling the position of ith row and j column of the matrix to construct a judgment matrix A ═ aij)m×mWherein a isij>0,aji＝1/aij. The decision matrix scaling method is shown in table 1.
TABLE 1 decision matrix Scale method
(3) And (5) checking the hierarchical single ordering and the consistency thereof.
The consistency check is needed to be carried out on the judgment matrix A, and consistency indexes of the difference degree of the judgment matrix and the consistency matrix are defined
There is complete consistency when CI is 0, there is more satisfactory consistency when CI is close to 0, the larger CI, the more severe the inconsistency, where m is the dimension of the matrix. In order to measure the CI size, a random consistency index RI is introduced, and according to the m size, an average random consistency index RI is searched according to a table 2.
TABLE 2 average random consistency index RI
Calculating a consistency ratio
If CR is<0.1, the consistency of the matrix is considered to be acceptable, if the constructed judgment matrix is too different from the consistency matrix, the judgment matrix is not used, and the judgment matrix A is reconstructed, namely, the judgment matrix A is aijAdjustments are made until a consistency check is passed. Let their weights be wt',t＝1,2,…m。
S1032: entropy weight method
The entropy weight method is an objective assignment method for determining the index weight according to the evaluation index matrix, is simple in calculation, can eliminate the subjectivity of each factor weight as much as possible, eliminates the influence of subjective factors, and can visually express the influence of each influencing factor on the index, so that the evaluation result is more practical. The calculation steps of the entropy method are as follows.
Suppose there are m evaluation index factors, where each index has n pieces of data, i.e., X ═ X1(k),x2(k),…xm(k) 1,2, … n, normalized to xijI is 1,2, … m; j is 1,2, … n. Information entropy e of a set of datajAs shown in formula (5).
In the formula:
the information entropy of each index is e1,e2,…emThen the weight coefficient
Is composed of
In the formula:
s1033: combination evaluation method
And a multiplier synthesis method is adopted to carry out weight combination on an analytic hierarchy process evaluation method and an entropy weight method, and the influence strength of each influence factor on the index is visually represented. And combining the entropy weight method and the analytic hierarchy process according to the formula to obtain the comprehensive weight.
In the formula: j is 1,2, …, m, w' is the analytic hierarchy weight, w*As entropy weight
Thus the comprehensive evaluation function is
F＝w1x1+w2x2+…+wmxm (8)
S104: establishing a slurry circulating pump operation optimization model by adopting an SVM classifier based on historical optimal operation data;
inputting historical operation data of the slurry circulating pump into an SVM classifier, and outputting a slurry circulating pump operation combination mode.
S105: inputting the real-time operation data of the slurry circulating pump into the slurry circulating pump operation optimization model, and outputting the slurry circulating pump operation suggestion combination scheme.
Example two
In this embodiment, the method provided in the first embodiment is applied to a power plant in the east of Shandong, and specific experimental results and analysis are described below:
the simulation experiment data is derived from 2017, 6-8 month operation data of a No. 1 unit of a power plant in Shandong, the sampling time interval is 5 minutes, 26496 pieces in total are obtained, and the data comprises 40 dimensions such as unit load, inlet sulfur dioxide concentration and the like. The unit has 6 slurry circulating pumps A-F, and the technical specifications are shown in Table 3. The 6 slurry circulating pump motors are 10000V three-phase motors, and the power calculation formula is
Wherein
The value of the power factor is generally between 0.7 and 0.85, so the power consumption of the slurry circulating pump can be represented by the current of the slurry circulating pump. And coding according to the running state, wherein the running state is 1, the non-running state is 0, and the binary coding of the running state of the slurry circulating pump can be obtained.
TABLE 3 slurry circulating pump specification
In order to improve the data quality, 24275 data are obtained after preprocessing the acquired data samples. And preliminarily selecting 17-dimensional data to describe the operation characteristics of the slurry circulating pump by combining the desulphurization process mechanism analysis and the production experience. And finally selecting 7-dimensional data of the unit load, the inlet sulfur dioxide concentration, the discharged flue gas flow velocity, the outlet pressure of the absorption tower, the inlet flue gas converted flow of the absorption tower, the outlet flue gas converted flow of the absorption tower and the inlet pressure of the absorption tower as main factors influencing the operation of the slurry circulating pump through mutual information and correlation coefficient quantitative analysis.
1.1 division of operating modes
And clustering the unit operation conditions based on the unit load and the inlet sulfur dioxide concentration by adopting an FCM algorithm, and performing cross combination on clustering results to obtain uniform distribution of each load section and each inlet sulfur dioxide section.
The best clustering result can be obtained when the unit load is clustered into 7 classes according to the Xie-Beni index, and the data distribution is shown in FIG. 2. In 7 working conditions of unit load division, the load distribution difference is obvious, the minimum load interval is working condition 4, the load interval is 72.31MW, the maximum load interval is working condition 1, and the interval is 85.5 MW. According to the actual operation condition, the main operation load of the power plant is more than 600MW, the data are less under the working condition that the working condition is lower than 600MW in the working condition division and other working conditions, and the division span of the distribution interval is larger, so that the division result can reasonably reflect the load working condition division of the unit.
The best clustering results were obtained when the inlet sulfur dioxide concentration was clustered to class 9, the data distribution of which is shown in fig. 3. The maximum dividing interval of the working condition 1 is 446.3mg/m3, and the minimum dividing interval of the working condition 4 is 177.8mg/m 3; in actual operation, the concentration of sulfur dioxide is mainly about 1400mg/m3, the division is more accurate in the working condition of a data dense area in the division process, and the interval is smaller. When the concentration of the sulfur dioxide is less than 1000mg/m3 or more than 2200mg/m3, the data distribution is less, and the range is properly enlarged when the working condition is divided. Therefore, the dividing result can reasonably reflect the working condition division of the sulfur dioxide at the inlet.
The method breaks through the limitation of uniform division in the original division method, and the obtained different clustering working condition data are scientifically distributed in the working condition intervals of unit load and inlet sulfur dioxide reduced concentration. And in the construction of the target working condition library, the same type of optimal data is easily extracted from each divided working condition to construct the optimal target working condition library. The two operating conditions were cross-combined to obtain 63 combined conditions, and the results of the condition division are shown in table 4.
TABLE 4 Combined operating conditions
1.2 historical optimal operating conditions acquisition
The historical operation data of the slurry circulating pump under each working condition cluster after the working conditions are divided is automatically optimized by using a combined evaluation method, and a hierarchical structure model is created as shown in fig. 4.
Constructing a judgment matrix according to the hierarchical structure model as shown in Table 5, and carrying out consistency check on the judgment matrix constructed in the Table 5 to judge the rationality of the judgment matrix, wherein the judgment matrix is a positive and reciprocal matrix and the maximum eigenvalue lambda of the judgment matrix ismax7.5750, the corresponding eigenvectors are normalized to
W＝(0.5530,0.7161,0.2167,0.3247,0.1225,0.0686,0.0962)T
Consistency index for judging difference degree between matrix and consistency matrix
And because the matrix size is 7 × 7, the average random consistency index RI is 1.36, and according to the consistency index CIAnd RI, calculating the consistency ratio
The consistency of the decision matrix may be deemed acceptable.
Performing weight analysis by using an analytic hierarchy process, and obtaining w 'of each influence factor weight according to the judgment matrix and the operation data'1＝0.2636,
w′2＝0.3414,w′3＝0.1033,w′4＝0.1548,w′5＝0.0584,
w′6＝0.0327,w′7＝0.0459
Using an entropy weight method to carry out weight analysis, and calculating the information entropy of each influence factor according to the operation data to obtain e1＝0.9986,e2＝0.9968,e3＝0.9992,e4＝0.9966,e5＝0.9992,e6＝0.9973,e70.9992, the weighting factor is
TABLE 5 decision matrix
Performing subjective and objective combination weight by using multiplier synthesis method to obtain comprehensive weight w1＝0.1662,w2＝0.4878,w3＝0.0360,
w4＝0.2343,w5＝0.0204,w6＝0.0394,w70.0159 thus the overall evaluation function is:
F＝0.1662x1+0.4878x2+0.0360x3+0.2343x4+
0.0204x5+0.0394x6+0.0159x7
in the formula: x is the number of1Load of unit, x2Inlet sulfur dioxide reduced concentration, x3Conversion flow rate x of flue gas at inlet of absorption tower4Absorption column inlet pressure, x5Conversion flow rate x of flue gas at outlet of absorption tower6Outlet pressure, x, of the absorption column7The flow rate of the discharged flue gas.
The combined evaluation coefficient is an evaluation value generated under the manual operation condition of the slurry circulating pump, the difference value between the evaluation index value and the actual operation value is calculated, and the data when the difference value is larger is taken as the optimal operation record in the historical operation process. And automatically optimizing historical operating data of the slurry circulating pump by using a combined evaluation function in each working condition cluster, converging the optimal operating data in each working condition, and finally obtaining 2422 pieces of historical optimal working condition data which are used as the operating optimization reference of the slurry circulating pump in the desulfurization process of the power plant. Some of the data are shown in table 6.
TABLE 6 optimal operating conditions data
1.3 SVM classifier
An SVM classifier model is used, 7 main influence factors influencing power consumption of a slurry circulating pump are used as input, a slurry circulating pump operation combination mode code is used as output, a genetic algorithm is used for carrying out optimization training classification model on two parameters C and g of the SVM, 2422 data in an optimal working condition library are divided into a training set and a testing set according to the proportion of 8:2 for carrying out experiments, the accuracy under 5-fold cross validation is used as a genetic algorithm fitness function, and the optimal fitness and the average fitness of the genetic algorithm change along with evolution algebra as shown in figure 5.
The optimal fitness and the average fitness of the population gradually tend to be stable along with the increase of evolution generations, and slowly converge when evolving to the 7 th generation, the optimal fitness value is 96.69%, the obtained optimal penalty coefficient C is 29.2858, and the kernel function g is 9.0969. The model is used for carrying out classification test on the test set, the classification accuracy is 95.59%, and the classification model is higher in classification accuracy and higher in applicability when used in a slurry circulating pump operation mode.
In order to test the energy-saving and consumption-reducing effects of the actual operation of the proposed method, 600 pieces of operation record simulation power plant real-time operation data are randomly selected from the actual operation data of the power plant in 6-8 months to verify the consumption-reducing effects of the proposed method. 600 pieces of simulation data are input into a classifier, and the operation mode and the actual operation mode of the slurry circulating pump, the recommended operation current and the actual rated current are given as a pair ratio shown in the table 7. According to the actual operation condition of desulfurization, under the condition that the load of the unit is the same, the concentration of sulfur dioxide at the inlet and the SO2 emission meeting the environmental protection requirement, the lower the current of the slurry circulating pump, the more the optimal operation is met. Under the conditions of the same unit load, the reduced concentration of the sulfur dioxide at the inlet and the emission of SO2 meeting the environmental protection requirement, the lower the current of the slurry circulating pump, the more the optimal operation is met. Actual operation conditions, measurement errors and other factors in the desulfurization operation, and the actual operation current of the slurry circulating pump fluctuates, so the total current of the recommended operation mode is the average current of the slurry circulating pump in the recommended operation state.
Through analysis of operation data of 600 optimized slurry circulating pumps, 426 slurry circulating pump power consumption data are reduced through operation optimization, and the operation data account for 71.00% of operation optimization total data such as operation data with the sequence numbers of 1-3, 5 and 8-11 in the table 7; the running optimization results and the original running mode are consistent, and the running optimization results and the original running mode have 149 data, such as running data with the sequence number 4 and the sequence number 7 in the table 7, which account for 24.83% of running optimization total data; the power consumption is higher than that of 25 pieces of operation optimization data which are not realized in the original combination mode, such as serial number 6 operation data in table 7, and accounts for 4.17% of the operation optimization total data. In the operation optimization, the power consumption of the slurry circulating pump combined operation mode is suggested to be reduced by about 21.55 percent on average, and a comparison graph of the optimized power consumption is shown in fig. 6.
The slurry circulating pump is a main energy consumption device of a desulfurization system of a thermal power plant, the operation combined mode of the slurry circulating pump is mainly controlled in a manual operation mode at present, and the slurry circulating pump has great operation optimization potential. In this embodiment, based on a large amount of historical operating data accumulated by a plant-level monitoring information platform of a power plant, a method based on the combination of clustering and classification is provided to realize the operation optimization of the slurry circulating pump: (1) and providing a working condition division method for clustering and dividing the unit load and the inlet sulfur dioxide concentration respectively through FCM and then performing cross combination. By analyzing the working condition division result, the method accords with the practical production. (2) And automatically acquiring the optimal historical operating data of the slurry circulating pump by adopting a combined evaluation method based on an analytic hierarchy process and an entropy weight method for each working condition data. (3) And establishing a slurry circulating pump operation optimization model by using historical optimal working condition data and adopting an SVM classifier based on a genetic algorithm. Based on the measured data of the power plant, simulation experiments show that: the operation optimization method for the slurry circulating pump can reduce the power consumption of the slurry circulating pump by about 21.55% to the maximum extent, is feasible based on the data-driven operation optimization method for the slurry circulating pump, and can further develop engineering realization research.
EXAMPLE III
The implementation provides a slurry circulating pump operation optimization system based on data driving.
A data-driven slurry circulation pump operation optimization system, comprising:
a data acquisition module configured to: acquiring historical operation data of a slurry circulating pump;
a clustering module configured to: clustering the unit operation conditions based on historical operation data of the slurry circulating pump by adopting an FCM algorithm, and performing cross combination on clustering results to obtain the condition data of different operation combinations;
a comprehensive evaluation module configured to: automatically acquiring historical optimal operation data in the working condition data by adopting a comprehensive evaluation function;
a model building module configured to: establishing a slurry circulating pump operation optimization model by adopting an SVM classifier based on historical optimal operation data;
an output module configured to: inputting the real-time operation data of the slurry circulating pump into the slurry circulating pump operation optimization model, and outputting the slurry circulating pump operation suggestion combination scheme.
It should be noted here that the data obtaining module, the clustering module, the comprehensive evaluation module, the model building module, and the output module correspond to steps S101 to S105 in the first embodiment, and the modules are the same as the corresponding steps in the implementation example and the application scenario, but are not limited to the disclosure in the first embodiment. The specific implementation process of this embodiment is the same as that of the embodiment, but is not limited to the disclosure of the embodiment two.
Example four
The present embodiment provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps in the data-driven-based slurry circulation pump operation optimization method according to the first embodiment.
EXAMPLE five
The present embodiment provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the steps of the data-driven slurry circulation pump operation optimization method according to the embodiment.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A method for optimizing operation of a slurry circulating pump based on data driving is characterized by comprising the following steps:
acquiring historical operation data of a slurry circulating pump;
clustering the unit operation conditions based on historical operation data of the slurry circulating pump by adopting an FCM algorithm, and performing cross combination on clustering results to obtain the condition data of different operation combinations;
automatically acquiring historical optimal operation data in the working condition data by adopting a comprehensive evaluation function;
establishing a slurry circulating pump operation optimization model by adopting an SVM classifier based on historical optimal operation data;
inputting the real-time operation data of the slurry circulating pump into the slurry circulating pump operation optimization model, and outputting the slurry circulating pump operation suggestion combination scheme.
2. The method for optimizing the operation of the slurry circulating pump based on the data driving of claim 1, wherein the construction process of the comprehensive evaluation function comprises the following steps:
based on an analytic hierarchy process and an entropy weight method, a multiplier synthesis method is adopted to construct a comprehensive evaluation function.
3. The method for optimizing the operation of the slurry circulating pump based on the data driving of claim 1, wherein the historical operation data of the slurry circulating pump comprises unit load operation data, inlet sulfur dioxide operation data, absorption tower inlet flue gas reduced flow, absorption tower inlet pressure, absorption tower outlet flue gas reduced flow, absorption tower outlet pressure and exhaust flue gas flow rate.
4. The method for optimizing the operation of a slurry circulating pump based on data driving according to claim 1, wherein the working condition data of different operation combinations comprises: operating condition data for each load section and each inlet sulfur dioxide section.
5. The method for optimizing the operation of the slurry circulating pump based on data driving according to claim 1, wherein Xie-Beni indexes are adopted to evaluate clustering results and determine the number of clustered clusters.
6. The method for optimizing the operation of the slurry circulating pump based on data driving of claim 1, wherein historical operation data of the slurry circulating pump is input into an SVM classifier, and a combination mode of operation of the slurry circulating pump is output.
7. The method for optimizing the operation of the slurry circulating pump based on data driving according to claim 1, wherein after obtaining the historical operation data of the slurry circulating pump, the method comprises the following steps: and preprocessing the historical operating data and eliminating abnormal data.
8. A data-driven slurry circulation pump operation optimization system, comprising:
a data acquisition module configured to: acquiring historical operation data of a slurry circulating pump;
a clustering module configured to: clustering the unit operation conditions based on historical operation data of the slurry circulating pump by adopting an FCM algorithm, and performing cross combination on clustering results to obtain the condition data of different operation combinations;
a comprehensive evaluation module configured to: automatically acquiring historical optimal operation data in the working condition data by adopting a comprehensive evaluation function;
a model building module configured to: establishing a slurry circulating pump operation optimization model by adopting an SVM classifier based on historical optimal operation data;
an output module configured to: inputting the real-time operation data of the slurry circulating pump into the slurry circulating pump operation optimization model, and outputting the slurry circulating pump operation suggestion combination scheme.
9. A computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, carries out the steps of the data-driven based slurry circulation pump operation optimization method according to any one of claims 1 to 7.
10. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the steps in the data-driven based slurry circulation pump operation optimization method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110368339.7A CN113094988A (en) | 2021-04-06 | 2021-04-06 | Data-driven slurry circulating pump operation optimization method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110368339.7A CN113094988A (en) | 2021-04-06 | 2021-04-06 | Data-driven slurry circulating pump operation optimization method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113094988A true CN113094988A (en) | 2021-07-09 |
Family
ID=76674392
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110368339.7A Pending CN113094988A (en) | 2021-04-06 | 2021-04-06 | Data-driven slurry circulating pump operation optimization method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113094988A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113669249A (en) * | 2021-08-27 | 2021-11-19 | 福建龙净环保股份有限公司 | Method, device and equipment for realizing selection of circulating pump |
| CN114607616A (en) * | 2022-04-12 | 2022-06-10 | 重庆远达烟气治理特许经营有限公司科技分公司 | Energy consumption diagnosis method and system for slurry circulating pump based on desulfurization operation history |
| CN114749006A (en) * | 2022-03-28 | 2022-07-15 | 浙江浙能兰溪发电有限责任公司 | Method for optimizing wet desulphurization slurry circulating pump |
| CN114862074A (en) * | 2022-07-11 | 2022-08-05 | 大唐环境产业集团股份有限公司 | Operation combination optimization method for slurry circulating pump of desulfurization system |
| CN116236892A (en) * | 2023-03-16 | 2023-06-09 | 福建龙净环保股份有限公司 | Method and device for controlling desulfurization equipment |
| CN116440670A (en) * | 2023-04-12 | 2023-07-18 | 华能伊春热电有限公司 | Limestone slurry density stability control method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109584611A (en) * | 2018-12-17 | 2019-04-05 | 山东建筑大学 | Parking lot free time parking stall judgment means and method based on ERI and network |
| CN111275335A (en) * | 2020-01-20 | 2020-06-12 | 华电莱州发电有限公司 | Data-driven slurry circulating pump optimization method and system |
-
2021
- 2021-04-06 CN CN202110368339.7A patent/CN113094988A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109584611A (en) * | 2018-12-17 | 2019-04-05 | 山东建筑大学 | Parking lot free time parking stall judgment means and method based on ERI and network |
| CN111275335A (en) * | 2020-01-20 | 2020-06-12 | 华电莱州发电有限公司 | Data-driven slurry circulating pump optimization method and system |
Non-Patent Citations (1)
| Title |
|---|
| 刘文慧: "基于数据驱动的火电厂脱硫浆液循环泵运行优化方法研究", 《中国优秀硕士学位论文全文数据库工程科技Ⅰ辑》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113669249A (en) * | 2021-08-27 | 2021-11-19 | 福建龙净环保股份有限公司 | Method, device and equipment for realizing selection of circulating pump |
| CN113669249B (en) * | 2021-08-27 | 2023-09-12 | 福建龙净环保股份有限公司 | Method, device and equipment for realizing circulating pump selection |
| CN114749006A (en) * | 2022-03-28 | 2022-07-15 | 浙江浙能兰溪发电有限责任公司 | Method for optimizing wet desulphurization slurry circulating pump |
| CN114749006B (en) * | 2022-03-28 | 2023-08-15 | 浙江浙能兰溪发电有限责任公司 | Optimization method for wet desulfurization slurry circulating pump |
| CN114607616A (en) * | 2022-04-12 | 2022-06-10 | 重庆远达烟气治理特许经营有限公司科技分公司 | Energy consumption diagnosis method and system for slurry circulating pump based on desulfurization operation history |
| CN114862074A (en) * | 2022-07-11 | 2022-08-05 | 大唐环境产业集团股份有限公司 | Operation combination optimization method for slurry circulating pump of desulfurization system |
| CN114862074B (en) * | 2022-07-11 | 2022-09-23 | 大唐环境产业集团股份有限公司 | Operation combination optimization method for slurry circulating pump of desulfurization system |
| CN116236892A (en) * | 2023-03-16 | 2023-06-09 | 福建龙净环保股份有限公司 | Method and device for controlling desulfurization equipment |
| CN116440670A (en) * | 2023-04-12 | 2023-07-18 | 华能伊春热电有限公司 | Limestone slurry density stability control method |
| CN116440670B (en) * | 2023-04-12 | 2023-10-31 | 华能伊春热电有限公司 | Limestone slurry density stability control method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113094988A (en) | Data-driven slurry circulating pump operation optimization method and system | |
| CN112069567B (en) | Method for predicting compressive strength of concrete based on random forest and intelligent algorithm | |
| CN110263988B (en) | Data operation optimization method based on power plant desulfurization system | |
| CN109634121B (en) | Multi-parent genetic algorithm air source heat pump multi-objective optimization control method based on radial basis function neural network | |
| CN110544011B (en) | Intelligent system combat effectiveness evaluation and optimization method | |
| CN112990556A (en) | User power consumption prediction method based on Prophet-LSTM model | |
| CN110571792A (en) | Analysis and evaluation method and system for operation state of power grid regulation and control system | |
| CN103745273A (en) | Semiconductor fabrication process multi-performance prediction method | |
| CN112364560B (en) | Intelligent prediction method for working hours of mine rock drilling equipment | |
| CN114757104B (en) | Method for constructing hydraulic real-time regulation and control model of series gate group water transfer project | |
| CN112070356A (en) | Method for predicting anti-carbonization performance of concrete based on RF-LSSVM model | |
| CN111476274B (en) | Big data predictive analysis method, system, device and storage medium | |
| CN115049019B (en) | Method and device for evaluating arsenic adsorption performance of metal organic framework and related equipment | |
| CN115470862A (en) | Dynamic self-adaptive load prediction model combination method | |
| CN112465239A (en) | Desulfurization system operation optimization method based on improved PSO-FCM algorithm | |
| CN112765902A (en) | RBF neural network soft measurement modeling method based on TentFWA-GD and application thereof | |
| CN104732067A (en) | Industrial process modeling forecasting method oriented at flow object | |
| CN115860173A (en) | Construction and prediction method and medium of carbon emission prediction model based on Stacking algorithm | |
| CN111815039A (en) | Weekly scale wind power probability prediction method and system based on weather classification | |
| Gao et al. | Fault diagnosis for cooling dehumidifier based on fuzzy classifier optimized by adaptive genetic algorithm | |
| CN113762591A (en) | Short-term electric quantity prediction method and system based on GRU and multi-core SVM counterstudy | |
| CN104239046B (en) | Software adaptive approach based on HMM and multi-objective Evolutionary Algorithm | |
| CN112070200A (en) | Harmonic group optimization method and application thereof | |
| CN111967689A (en) | Model and method for wind power generation prediction by combining multivariate and stepwise linear regression and artificial neural network | |
| CN113379093A (en) | Energy consumption analysis and optimization method for oil gas gathering and transportation system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |