CN113065772B - Method for constructing heating system regulation and control target model based on k-means clustering algorithm - Google Patents
Method for constructing heating system regulation and control target model based on k-means clustering algorithm Download PDFInfo
- Publication number
- CN113065772B CN113065772B CN202110366770.8A CN202110366770A CN113065772B CN 113065772 B CN113065772 B CN 113065772B CN 202110366770 A CN202110366770 A CN 202110366770A CN 113065772 B CN113065772 B CN 113065772B
- Authority
- CN
- China
- Prior art keywords
- data
- target
- temperature
- working condition
- water supply
- 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.)
- Active
Links
- 238000010438 heat treatment Methods 0.000 title claims abstract description 45
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000003064 k means clustering Methods 0.000 title claims abstract description 19
- 230000002159 abnormal effect Effects 0.000 claims abstract description 20
- 238000004140 cleaning Methods 0.000 claims abstract description 12
- 230000001105 regulatory effect Effects 0.000 claims abstract description 8
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 88
- 239000013598 vector Substances 0.000 claims description 15
- 238000005286 illumination Methods 0.000 claims description 11
- 230000005855 radiation Effects 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 7
- 238000003032 molecular docking Methods 0.000 claims description 6
- 230000001276 controlling effect Effects 0.000 claims description 4
- 210000001503 joint Anatomy 0.000 claims description 4
- 235000012571 Ficus glomerata Nutrition 0.000 claims description 3
- 244000153665 Ficus glomerata Species 0.000 claims description 3
- 230000001502 supplementing effect Effects 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 238000013499 data model Methods 0.000 claims description 2
- 230000035772 mutation Effects 0.000 claims 1
- 238000005259 measurement Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- 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
- G06Q10/067—Enterprise or organisation modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Operations Research (AREA)
- Evolutionary Computation (AREA)
- Development Economics (AREA)
- Evolutionary Biology (AREA)
- Educational Administration (AREA)
- Bioinformatics & Computational Biology (AREA)
- Game Theory and Decision Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Air Conditioning Control Device (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses a method for constructing a heating system regulation and control target model based on a k-means clustering algorithm, which comprises the following steps: s1, establishing a target model library regulated and controlled by a heating system; s2, butting a heat supply system SCADA system database to obtain operation data, wherein the operation data comprises a heat source, heat station operation data, working condition data and basic information data; s3, based on an algorithm, performing data cleaning on the operation data in the step S2, and eliminating abnormal data; and S4, selecting a regulation target, and constructing a target model regulated by the heating system based on a k-means clustering algorithm. The invention provides a method for constructing a heating system regulation target model based on a k-means clustering algorithm, which is used for constructing more accurate and refined regulation target models on a heat source side and a heating station side, realizing refined regulation and control as required on a heating system, improving the operation economy of the heating system and reducing the overload.
Description
Technical Field
The invention relates to a method for constructing a heating system regulation and control target model based on a k-means clustering algorithm, and belongs to the field of heating predictive control.
Background
At present, a heating system is a typical large-lag and strong-coupling system, and for the regulation and control of the heating system, at present, much more operation experience of a dispatcher is relied on, wherein the regulation and control based on temperature is the most common mode used in the heating system, because the temperature perceptibility is strong, the method has operability and strong practicability, however, the main problems are as follows: the method comprises the steps that an operator determines the target temperature of each heating station according to experience to regulate, station data are numerous, the heating target temperature of each heating station is related to factors such as the type of a building, the heat preservation characteristic and the heating form of a community, because the influence factors are more, the characteristics of the communities are different, the delay phenomenon at the same moment is different, the relation among the factors is difficult to grasp manually and accurately, and therefore the target temperature model of each station is given accurately. Moreover, regulation and control of the heat supply load at the heat source side are mostly determined by people according to heat indexes through experience, but the structure of the heat supply system is updated quickly every year, the experience accumulation is slow, and certain deviation exists, so that waste of the heat supply load is caused, and meanwhile, the heat supply system is not beneficial to implementing more refined regulation and control.
Disclosure of Invention
The invention aims to solve the technical problem of overcoming the defects of the prior art and provides a method for constructing a regulation and control target model of a heat supply system based on a k-means clustering algorithm, so that more accurate and refined regulation and control target models are constructed on a heat source side and a heat station side, the refined regulation and control according to needs of the heat supply system are realized, the operation economy of the heat supply system is improved, and the overload is reduced.
In order to solve the technical problems, the technical scheme of the invention is as follows:
a method for constructing a heating system regulation and control target model based on a k-means clustering algorithm comprises the following steps:
s1, establishing a target model library regulated and controlled by a heating system;
s2, butting a heat supply system SCADA system database to obtain operation data, wherein the operation data comprises a heat source, heat station operation data, working condition data and basic information data;
s3, based on an algorithm, performing data cleaning on the operation data in the step S2, and eliminating abnormal data;
and S4, selecting a regulation target, and constructing a target model for regulating and controlling the heating system based on a k-means clustering algorithm.
Further, the establishing of the target model library for heating system regulation in step S1 includes:
target bank GE = [ G, E ], including: a target type G and an influencing factor E, where G = [ G1, G2.. G, gi.. Gn ], n is a target number, and then Gi corresponds to the influencing factor E: ei = [ E1i, E2i. ·, eji. ·, E. The concrete expression is as follows:
s11, target type G1
The target is as follows: secondary side water supply temperature Tgss;
influencing factor E1: working condition data U (outdoor weather temperature To, humidity S, wind speed v, illumination Z and radiation W), primary network water supply temperature Tps, primary network backwater temperature Tpr, site flow Qpq, site area A and secondary network flow Qsq;
s12, target type G2
Target: the secondary side water supply average temperature Tgsa =1/2 (Tss + Tsr);
influencing factor E2: working condition data U (outdoor weather temperature To, humidity S, wind speed v, illumination Z and radiation W), primary network water supply temperature Tps, primary network backwater temperature Tpr, site flow Qpq and site area A;
s13, target type G3
Target: secondary side return water temperature Tgsr;
influencing factor E3: working condition data U (outdoor weather temperature To, humidity S, wind speed v, illumination Z and radiation W), primary network water supply temperature Tps, primary network backwater temperature Tpr, site flow Qpq, site area A and secondary network flow Qsq;
s14, target type G4
The target is as follows: valve opening Kgs;
influencing factor E4: the system comprises a heat source side water supply temperature Tss, a water supply flow Qs, a water supply pressure Pss, a water return pressure Psr, working condition data U (outdoor weather temperature To, humidity S, wind speed v, illumination Z and radiation W), a primary network water supply temperature Tps, a primary network water return temperature Tpr, a site flow Qpq, a primary network water supply pressure Psite area A, a secondary network water supply temperature Tss, a secondary network water return temperature Tsr, a secondary network flow Qsq, a primary network water supply pressure Pps, a primary network water return pressure Ppr and a valve opening Ksh;
s15, target type G5
The target is as follows: station traffic Qgsq;
influencing factor E5: working condition data U (outdoor weather temperature To, humidity S, wind speed v, illumination Z and radiation W), primary network water supply temperature Tps, primary network backwater temperature Tpr, site flow Qpq, site area A and secondary network flow Qsq.
Further, the step S2 of docking the SCADA system database of the heat supply system to obtain the heat source, the operation data of the thermal station, the working condition data, and the basic information data includes:
the SCADA system is used for acquiring historical operation data of the heat supply system and then storing the historical operation data into the database.
The specific data attributes of the operating data include: the method comprises the following steps of (1) supplying water at a heat source side by a temperature Tss, a water supply flow Qs, a water supply pressure Pss and a return water pressure Psr; the method comprises the following steps that the primary network water supply temperature Tps, the primary network water return temperature Tpr, the station flow Qpq, the primary network water supply pressure Pps, the primary network water return pressure Ppr, the station area A, the secondary network water supply temperature Tss, the secondary network water return temperature Tsr and the secondary network flow Qsq of each heating station are obtained; working condition data U and indoor temperature Ti; the basic data includes: the length L of each pipe section and the data D of each pipe diameter are as follows:
s21, heat source data protocol
S22, thermal power station data protocol
S23, basic data information
S24, working condition data docking protocol
Further, in step S3, based on the algorithm, data cleaning is performed on the operating data to remove abnormal data, including:
defining the total amount of sample data of the butt joint as m, and expressing a data set as follows: x = [ X1, X2.,. Once, xi ], xm ], where Xi = [ X1i, X2i.. Xji.. X yi ], xij is the jth attribute value of the ith data;
the data attribute of the sample data comprises heat supply target data Gi = [ Gi1, gi2,. Multidot., gij.. Multidot., gik ] and corresponding influence factors Ei = [ Ei1, ei2,. Multidot., eij., eik ], and y is the total number of the data attribute;
for the missing data, the following processing is carried out according to the rule:
if the current piece of data is Xi = [ x1i, x2i.. Xij.. Xyz ], then the missing data is cleaned according to the following steps:
s31, if the proportion of the number of the missing attributes of the current data record to the total number of the attributes of the data exceeds theta%, the following steps are carried out:
discarding, and automatically supplementing the number of 1 sample data after cleaning the data to meet the requirement of a data set;
s32, when the proportion is less than theta%, filling the data attribute, and enabling the missing attribute to be a discontinuous value x ij In time, the measurement time interval of the data of the heating system is short, generally tens of seconds or minutes, and if the previous data is adopted for filling, the attribute value x is lost ij ：
x ij ＝x i-1 ，j；
S33, when the missing proportion is smaller than theta% and the missing attribute is a continuous value, simplifying the attribute of the running data into a continuously changing process, and filling the data by adopting the arithmetic mean value of the previous data and the next data; then:
s34, aiming at abnormal data, namely, when the data mutates at a certain moment, replacement processing is also needed; judging abnormal data x ij The method comprises the following steps:
when in use
Replacing the abnormal value by adopting a method S32 and a method S33;
wherein:
l i : the number of missing attributes in the data Xi;
θ: judging whether the data is selected as the missing proportion threshold of the sample data;
x j * : the mean value of the attribute values of all sample data in the jth column;
beta: is a threshold value for judging whether the data attribute is abnormal or not.
Further, selecting a regulation target in the step S4, and constructing a target model for heating system regulation based on a k-means clustering algorithm, including:
selecting a regulation target Gi of a heat supply system, selecting a sample set of heat supply target model training data as X according to corresponding influence factor Ei data, and carrying out the following main processes:
s41, dividing the number of target cluster trees: dividing according to outdoor working conditions, and enabling the total amount to be k (the k value can be determined according to the difference value between the maximum temperature and the minimum temperature of the outdoor working conditions);
s42, inputting a working condition sample set: x = [ X1, X2,.. Multidot., xj.. Multidot., xm ], and the maximum number of iterations is a;
s43, randomly selecting k samples from the data set E as initial k centroid vectors: μ = [ μ 1, μ 2,.. Mu.i,. Mu.k ];
s44, for a =1,2.., a, determining a target model cluster according to the following steps:
s441, C is initialized to
When the current is over;
s442, for i =1,2.. M, calculating the distance of the sample Xj and each centroid vector μ j (j =1,2.. K): dij = | | Xi- μ j | |2, and marking Xi as the minimum type λ i corresponding to dij;
at this time, update is performed
S443, for j =1,2.. K, the new centroid is recalculated for all sample points in Ci:
s444, if all the centroid vectors are not changed, turning to the step S45;
s45, dividing a corresponding heat supply target model cluster: c = [ C1, C2,. Cndot, ci.., ck ] and the corresponding centroid vector μ = [ μ 1, μ 2,. Cndot, μ i,. Cndot ], which includes target values of: g × i = [ G × 1, G × 2., G × i., G × k ];
s46, acquiring weather forecast information data at the next moment based on the target model cluster, and realizing prediction of a next working condition data target;
and (4) making the working condition data be Xt, calculating the distance between the working condition data and each centroid vector, classifying the target cluster Ct, and correspondingly obtaining the target G x t under the working condition.
By adopting the technical scheme, the K-means algorithm is applied to prediction of different target models of the heat supply system through the operation data of the butt joint heat supply system, the target strategy of the operation regulation and control process of the heat supply system under the future working condition can be quickly obtained, the practicability is high, and the operation regulation and control level of the heat supply system can be effectively improved.
Drawings
FIG. 1 is a flow chart of a method for constructing a heating system regulation and control target model based on a k-means clustering algorithm.
Detailed Description
In order that the present invention may be more readily and clearly understood, a more particular description of the invention briefly described above will be rendered by reference to specific embodiments that are illustrated in the appended drawings.
As shown in fig. 1, a method for constructing a heating system regulation and control target model based on a k-means clustering algorithm comprises the following steps:
s1, establishing a target model library regulated and controlled by a heating system;
s2, butting a heat supply system SCADA system database to obtain operation data, wherein the operation data comprises heat sources, heating power station operation data, working condition data and basic information data;
s3, based on an algorithm, performing data cleaning on the operation data in the step S2, and eliminating abnormal data;
and S4, selecting a regulation target, and constructing a target model for regulating and controlling the heating system based on a k-means clustering algorithm.
As shown in fig. 1, the establishing of the target model library for heating system regulation in step S1 includes:
the control targets of different heating power companies are different, the control targets of all heat supply enterprises are supported to be generated, and a control target library is established. Target bank GE = [ G, E ], including: a target type G and an influencing factor E, where G = [ G1, G2.. G, gi.. Gn ], n is a target number, and then Gi corresponds to the influencing factor E: ei = [ E1i, E2i. ·, eji. ·, E. The concrete expression is as follows:
s11, target type G1
The target is as follows: secondary side water supply temperature Tgss;
influencing factor E1: working condition data U (outdoor weather temperature To, humidity S, wind speed v, illumination Z and radiation W), primary network water supply temperature Tps, primary network backwater temperature Tpr, site flow Qpq, site area A and secondary network flow Qsq;
s12, target type G2
The target is as follows: the secondary side supply and return water average temperature Tgsa =1/2 (Tss + Tsr);
influencing factor E2: working condition data U (outdoor weather temperature To, humidity S, wind speed v, illumination Z and radiation W), primary network water supply temperature Tps, primary network backwater temperature Tpr, site flow Qpq and site area A;
s13, target type G3
Target: secondary side return water temperature Tgsr;
influencing factor E3: working condition data U (outdoor weather temperature To, humidity S, wind speed v, illumination Z and radiation W), primary network water supply temperature Tps, primary network backwater temperature Tpr, site flow Qpq, site area A and secondary network flow Qsq;
s14, target type G4
Target: valve opening Kgs;
influencing factor E4: the system comprises a heat source side water supply temperature Tss, a water supply flow Qs, a water supply pressure Pss, a water return pressure Psr, working condition data U (outdoor weather temperature To, humidity S, wind speed v, illumination Z and radiation W), a primary network water supply temperature Tps, a primary network water return temperature Tpr, a site flow Qpq, a primary network water supply pressure Psite area A, a secondary network water supply temperature Tss, a secondary network water return temperature Tsr, a secondary network flow Qsq, a primary network water supply pressure Pps, a primary network water return pressure Ppr and a valve opening Ksh;
s15, target type G5
The target is as follows: station traffic Qgsq;
influencing factor E5: working condition data U (outdoor weather temperature To, humidity S, wind speed v, illumination Z and radiation W), primary network water supply temperature Tps, primary network backwater temperature Tpr, site flow Qpq, site area A and secondary network flow Qsq;
according to the requirements of heat supply enterprises, different regulation and control targets can be selected and adopted. And the working condition data is weather information data, the weather information data of the sample data is historical data, and the future weather information data of the target working condition is derived from weather forecast.
As shown in fig. 1, the step S2 of docking the SCADA system database of the heat supply system to obtain the operation data, the working condition data, and the basic information data of the heat source and the thermal station includes:
the SCADA system is used for acquiring historical operation data of the heat supply system, then storing the historical operation data into the database, generating a data model of each target in the target library based on a K-means algorithm and the operation data, and generating data of the heat supply system to be butted, wherein the data comprises the historical operation data of the target and corresponding influence factors of the target;
the specific data attributes of the operating data include: the method comprises the following steps of (1) supplying water at a heat source side by a temperature Tss, a water supply flow Qs, a water supply pressure Pss and a return water pressure Psr; the system comprises a primary network water supply temperature Tps, a primary network water return temperature Tpr, a station flow rate Qpq, a primary network water supply pressure Pps, a primary network water return pressure Ppr, a station area A, a secondary network water supply temperature Tss, a secondary network water return temperature Tsr and a secondary network flow rate Qsq of each thermal station; working condition data U and indoor temperature Ti; the basic data includes: the length L of each pipe section and the data D of each pipe diameter are as follows:
s21, heat source data protocol
S22, thermal power station data protocol
S23, basic data information
S24, working condition data docking protocol
As shown in fig. 1, in step S3, based on the algorithm, the data cleaning is performed on the operation data to remove abnormal data, including:
defining the total amount of sample data of the butt joint as m pieces, and expressing the data set as: x = [ X1, X2.,. Once, xi ], xm ], where Xi = [ X1i, X2i.. Xji.. X yi ], xij is the jth attribute value of the ith data;
the data attribute of the sample data comprises heat supply target data Gi = [ Gi1, gi2,. Multidot., gij.. Multidot., gik ] and corresponding influence factors Ei = [ Ei1, ei2,. Multidot., eij., eik ], and y is the total number of the data attribute;
in the data cleaning process, missing values and abnormal values are mainly processed, related data are filled, and the following processing is performed on the missing data according to rule processing:
if the current piece of data is Xi = [ x1i, x2i.. Xij.. Xyz ], then the missing data is cleaned according to the following steps:
s31, if the proportion of the number of the missing attributes of the current data record to the total number of the attributes of the data exceeds theta%, the following steps are carried out:
discarding, and automatically supplementing the number of 1 sample data after cleaning the data to meet the requirement of a data set;
s32, when the proportion is less than theta%, filling the data attribute, and enabling the missing attribute to be a discontinuous value x ij In time, the measurement time interval of the data of the heating system is short, generally tens of seconds or minutes, and if the previous data is adopted for filling, the attribute value x is lost ij ：
x ij ＝x i-1 ，j；
And S33, when the missing proportion is smaller than theta% and the missing attribute is a continuous value, simplifying the attribute of the running data into a continuously changing process, and filling the data by adopting the arithmetic mean value of the previous data and the next data. Then:
s34, replacement processing is also required for abnormal data, that is, data that suddenly changes at a certain time. Judging abnormal data x ij The method comprises the following steps:
when in use
Then, the abnormal value is replaced by using the method S32 and the method S33.
Wherein:
l i : the number of missing attributes in the data Xi;
θ: judging whether the data is selected as the missing proportion threshold of the sample data;
x j * the mean value of the attribute values of all sample data in the jth column;
beta: a threshold value for determining whether the data attribute is abnormal.
As shown in fig. 1, selecting a regulation target in step S4, and constructing a target model for heating system regulation based on a k-means clustering algorithm, including:
selecting a regulation target Gi of a heat supply system, selecting a sample set of heat supply target model training data as X according to corresponding influence factor Ei data, and carrying out the following main processes:
s41, dividing the number of target cluster trees: dividing according to outdoor working conditions, and enabling the total amount to be k (the k value can be determined according to the difference value between the maximum temperature and the minimum temperature of the outdoor working conditions);
s42, inputting a working condition sample set: x = [ X1, X2,.. Multidot., xj.. Multidot., xm ], and the maximum number of iterations is a;
s43, randomly selecting k samples from the data set E as initial k centroid vectors: μ = [ μ 1, μ 2,.. Mu.i,. Mu.k ];
s44, for a =1,2.., a, determining a target model cluster according to the following steps:
s441, C is initialized to
When the current is over;
s442, for i =1,2.. M, calculating the distance of the sample Xj and each centroid vector μ j (j =1,2.. K): dij = | | Xi- μ j | |2, and marking Xi as the minimum type λ i corresponding to dij;
at this time, update is performed
S443, for j =1,2.. K, the new centroid is recalculated for all sample points in Ci:
s444, if all the centroid vectors are not changed, turning to the step S45;
s45, dividing a corresponding heat supply target model cluster: c = [ C1, C2,.. Ci.., ck ] and corresponding centroid vector μ = [ μ 1, μ 2,..., μ i.,. μ k ], included target values are: g × i = [ G × 1, G × 2., G × i., G × k ];
s46, acquiring weather forecast information data at the next moment based on the target model cluster, and realizing prediction of a next working condition data target;
and (4) setting the working condition data as Xt, calculating the distance between the working condition data and each centroid vector, and classifying the target cluster Ct, wherein the target under the working condition is G x t correspondingly.
The technical problems, technical solutions and advantages of the present invention have been described in detail with reference to the above embodiments, and it should be understood that the above embodiments are merely exemplary and not intended to limit the present invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (4)
1. A method for constructing a heating system regulation and control target model based on a k-means clustering algorithm is characterized by comprising the following steps:
s1, establishing a target model library for regulating and controlling a heating system;
s2, butting a heat supply system SCADA system database to obtain operation data, wherein the operation data comprises a heat source, heat station operation data, working condition data and basic information data;
s3, based on an algorithm, performing data cleaning on the operation data in the step S2, and eliminating abnormal data;
s4, selecting a regulation target, and constructing a target model for regulating and controlling the heating system based on a k-means clustering algorithm;
the step S4 includes:
selecting a regulation target Gi of a heat supply system, selecting a sample set of heat supply target model training data as X according to corresponding influence factor Ei data, and carrying out the following main processes:
s41, dividing the number of target cluster trees: dividing according to outdoor working conditions, wherein the total amount is k, and the value of k is determined according to the difference value between the maximum temperature and the minimum temperature of the outdoor working conditions;
s42, inputting a working condition sample set: x = [ X1, X2, ·.. Ang, xj,. Ang, xm ], maximum number of iterations is a;
s43, randomly selecting k samples from the data set E as initial k centroid vectors: μ = [ μ 1, μ 2,.. Mu.i,. Mu.k ];
s44, for a =1,2.., a, determining a target model cluster according to the following steps:
s441, C is initialized to
When the current is over;
s442, for i =1,2.. M, calculating the distance of the sample Xj and each centroid vector μ j (j =1,2.. K): dij = | | Xi- μ j | |2, and marking Xi as the minimum type λ i corresponding to dij;
at this timeUpdating
S443, for j =1,2.. K, the new centroid is recalculated for all sample points in Ci:
s444, if all the centroid vectors are not changed, turning to the step S45;
s45, dividing a corresponding heat supply target model cluster: c = [ C1, C2,.. Ci.., ck ] and corresponding centroid vector μ = [ μ 1, μ 2,..., μ i.,. μ k ], included target values are: g × i = [ G × 1, G × 2., G × i., G × k ];
s46, acquiring weather forecast information data at the next moment based on the target model cluster, and realizing prediction of a next working condition data target;
and (4) making the working condition data be Xt, calculating the distance between the working condition data and each centroid vector, classifying the target cluster Ct, and correspondingly obtaining the target G x t under the working condition.
2. The method for constructing the heating system regulation and control target model based on the k-means clustering algorithm according to claim 1, wherein the step S1 of establishing the heating system regulation and control target model library comprises:
establishing a control target library according to different control targets of different heating power companies, wherein a target library GE = [ G, E ], and the method comprises the following steps: a target type G and an influencing factor E, where G = [ G1, G2.. G, gi.. Gn ], n is a target number, and then Gi corresponds to the influencing factor E: ei = [ E1i, E2i. ·, eji. ·, E. The concrete expression is as follows:
s11, target type G1
The target is as follows: secondary side water supply temperature Tgss;
influencing factor E1: working condition data U, primary network water supply temperature Tps, primary network return water temperature Tpr, site flow Qpq, site area A and secondary network flow Qsq;
s12, target type G2
Target: the secondary side supply and return water average temperature Tgsa =1/2 (Tss + Tsr);
influencing factor E2: working condition data U, primary network water supply temperature Tps, primary network water return temperature Tpr, station flow Qpq and station area A;
s13, target type G3
The target is as follows: secondary side return water temperature Tgsr;
influencing factor E3: working condition data U, primary network water supply temperature Tps, primary network return water temperature Tpr, site flow Qpq, site area A and secondary network flow Qsq;
s14, target type G4
The target is as follows: valve opening Kgs;
influencing factor E4: the system comprises a heat source side water supply temperature Tss, a water supply flow Qs, a water supply pressure Pss, a water return pressure Psr, working condition data U, a primary network water supply temperature Tps, a primary network water return temperature Tpr, a station flow Qpq, a primary network water supply pressure P, a station area A, a secondary network water supply temperature Tss, a secondary network water return temperature Tsr, a secondary network flow Qsq, a primary network water supply pressure Pps, a primary network water return pressure Ppr and a valve opening Ksh;
s15, target type G5
The target is as follows: station traffic Qgsq;
influencing factor E5: working condition data U, primary network water supply temperature Tps, primary network return water temperature Tpr, site flow Qpq, site area A and secondary network flow Qsq;
the working condition data U comprises outdoor weather temperature To, humidity S, wind speed v, illumination Z and radiation W.
3. The method for constructing the heating system regulation and control target model based on the k-means clustering algorithm as claimed in claim 1, wherein the step S2 of docking the SCADA system database of the heating system to obtain the heat source, the thermal station operation data, the working condition data and the basic information data comprises:
the SCADA system is used for acquiring historical operation data of the heat supply system, then storing the historical operation data into the database, generating a data model of each target in the target library based on a K-means algorithm and the operation data, and generating data of the heat supply system to be butted, wherein the data comprises the historical operation data of the target and corresponding influence factors of the target;
the specific data attributes of the operating data include: the method comprises the following steps of (1) supplying water at a heat source side by a temperature Tss, a water supply flow Qs, a water supply pressure Pss and a return water pressure Psr; the method comprises the following steps that the primary network water supply temperature Tps, the primary network water return temperature Tpr, the station flow Qpq, the primary network water supply pressure Pps, the primary network water return pressure Ppr, the station area A, the secondary network water supply temperature Tss, the secondary network water return temperature Tsr and the secondary network flow Qsq of each heating station are obtained; working condition data U and indoor temperature Ti; the basic data includes: the length L of each pipe section and the data D of each pipe diameter are as follows:
s21, a heat source data protocol;
s22, a data protocol of the heating power station;
s23, basic data information;
and S24, a working condition data docking protocol.
4. The method for constructing the heating system regulation and control target model based on the k-means clustering algorithm according to claim 1, wherein the step S3 is based on the algorithm, and the data cleaning is performed on the operation data to remove abnormal data, and comprises the following steps:
defining the total amount of sample data of the butt joint as m, and expressing a data set as follows: x = [ X1, X2.,. Once, xi ], xm ], where Xi = [ X1i, X2i.. Xji.. X yi ], xij is the jth attribute value of the ith data;
the data attribute of the sample data comprises heat supply target data Gi = [ Gi1, gi2,. Multidot., gij.. Multidot., gik ] and corresponding influence factors Ei = [ Ei1, ei2,. Multidot., eij., eik ], and y is the total number of the data attribute;
for the missing data, the following processing is carried out according to the rule:
and if the current sample data is Xi = [ x1i, x2i.. Xij.. Xyj ], cleaning missing data according to the following steps:
s31, if the proportion of the number of the missing attributes of the current data record to the total number of the attributes of the data exceeds theta%, the following steps are carried out:
discarding, and automatically supplementing 1 sample data after cleaning the data to meet the requirement of a data set;
s32, when the proportion is less than theta%, filling the data attribute, and enabling the missing attribute to be a discontinuous value x ij When the data is filled by the previous data, the attribute value x is lacked ij ：
x ij ＝x i-1，j ；
S33, when the missing proportion is smaller than theta% and the missing attribute is a continuous value, simplifying the attribute of the running data into a continuously changing process, and filling the data by adopting the arithmetic mean value of the previous data and the next data; then:
s34, aiming at abnormal data, namely, the data is subjected to mutation at a certain moment and also needs to be subjected to replacement processing; judging abnormal data x ij The method comprises the following steps:
when in use
Replacing the abnormal value by adopting a method S32 and a method S33;
wherein:
l i : the number of missing attributes in the data Xi;
θ: judging whether the data is selected as the missing proportion threshold of the sample data;
x j * : the mean value of the attribute values of all sample data in the jth column;
beta: is a threshold value for judging whether the data attribute is abnormal or not.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110366770.8A CN113065772B (en) | 2021-04-06 | 2021-04-06 | Method for constructing heating system regulation and control target model based on k-means clustering algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110366770.8A CN113065772B (en) | 2021-04-06 | 2021-04-06 | Method for constructing heating system regulation and control target model based on k-means clustering algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113065772A CN113065772A (en) | 2021-07-02 |
| CN113065772B true CN113065772B (en) | 2023-03-24 |
Family
ID=76566061
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110366770.8A Active CN113065772B (en) | 2021-04-06 | 2021-04-06 | Method for constructing heating system regulation and control target model based on k-means clustering algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113065772B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111275579A (en) * | 2020-02-21 | 2020-06-12 | 浙江大学 | Heat supply system thermal delay time identification method based on Pearson correlation coefficient and moving average method |
| CN111473407A (en) * | 2020-04-10 | 2020-07-31 | 浙江大学 | Model-based centralized heating system on-demand accurate regulation and control method |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106253350B (en) * | 2016-08-11 | 2019-03-05 | 清华大学 | Thermo-electrically Unit erriger combined method based on heating network heat accumulation benefit |
| US10825113B2 (en) * | 2018-03-16 | 2020-11-03 | Dalian University Of Technology | Method for short-term generation scheduling of cascade hydropower plants coupling cluster analysis and decision tree |
| CN110332605B (en) * | 2019-07-16 | 2020-12-29 | 常州英集动力科技有限公司 | BP neural network-based heat supply network energy consumption index analysis and early warning method |
| CN111652422A (en) * | 2020-05-27 | 2020-09-11 | 常州信息职业技术学院 | Heat supply system load prediction method, device and system based on building classification |
-
2021
- 2021-04-06 CN CN202110366770.8A patent/CN113065772B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111275579A (en) * | 2020-02-21 | 2020-06-12 | 浙江大学 | Heat supply system thermal delay time identification method based on Pearson correlation coefficient and moving average method |
| CN111473407A (en) * | 2020-04-10 | 2020-07-31 | 浙江大学 | Model-based centralized heating system on-demand accurate regulation and control method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113065772A (en) | 2021-07-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113657031B (en) | Heat supply dispatching automation realization method, system and platform based on digital twin | |
| CN111256294B (en) | Model prediction-based optimization control method for combined operation of water chilling unit | |
| CN110513767B (en) | Heat supply network hydraulic balance regulation and control method and system based on resistance characteristics of heating power station | |
| CN109270842B (en) | Bayesian network-based regional heat supply model prediction control system and method | |
| CN108916986B (en) | Information physical fusion secondary pipe network variable flow hydraulic balance regulation and control method and system | |
| CN104865830B (en) | Dual-intelligent-optimization control method for unit load | |
| CN111580382B (en) | Unit-level heat supply adjusting method and system based on artificial intelligence | |
| CN111561732B (en) | Heat exchange station heat supply adjusting method and system based on artificial intelligence | |
| CN103591637A (en) | Centralized heating secondary network operation adjustment method | |
| CN110376895B (en) | Thermal power generating unit coordination control method based on hierarchical limited predictive control | |
| CN115823706B (en) | Self-adaptive variable pressure difference energy-saving control system and method for primary pump | |
| CN110400018B (en) | Operation control method, system and device for coal-fired power plant pulverizing system | |
| CN106703904A (en) | Method for optimizing steam distribution curves of steam turbines on basis of data mining technologies | |
| CN113065772B (en) | Method for constructing heating system regulation and control target model based on k-means clustering algorithm | |
| CN116436033A (en) | Temperature control load frequency response control method based on user satisfaction and reinforcement learning | |
| CN113110056A (en) | Heat supply intelligent decision-making method and intelligent decision-making machine based on artificial intelligence | |
| CN110517460A (en) | A kind of induced joint concrete temperature state interval prediction and warning method | |
| CN110880773A (en) | Power grid frequency modulation control method based on combination of data driving and physical model driving | |
| CN113222231A (en) | Wisdom heating system based on internet of things | |
| CN111062576B (en) | Pressure management system based on DMA partition least adverse point | |
| CN117575341A (en) | Gas flow scheduling evaluation method based on big data natural gas transmission network | |
| CN116646936A (en) | New energy consumption risk pre-control method and storage medium | |
| WO2006003446A2 (en) | Process-related systems and methods | |
| CN116231749A (en) | New energy power system dispatching method based on digital twin | |
| CN116324634A (en) | Method for operating a process engineering plant, process engineering plant and method for retrofitting a process engineering plant |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |