CN115579072A - Intelligent water quality monitoring control system and method for heat supply pipe network - Google Patents
Intelligent water quality monitoring control system and method for heat supply pipe network Download PDFInfo
- Publication number
- CN115579072A CN115579072A CN202211221992.1A CN202211221992A CN115579072A CN 115579072 A CN115579072 A CN 115579072A CN 202211221992 A CN202211221992 A CN 202211221992A CN 115579072 A CN115579072 A CN 115579072A
- Authority
- CN
- China
- Prior art keywords
- pipe network
- dosing
- water quality
- model
- monitoring
- 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
Abstract
The invention provides a system and a method for intelligently monitoring and controlling water quality of a heat supply pipe network, which comprises the following steps of S1: monitoring the water quality of the heat supply pipe network; s2: uploading water quality monitoring data of a pipe network; s3: judging the water quality abnormity of the pipe network; s4: constructing and optimizing a dosing model; s5: automatically checking a table by a program to determine the dosing weight of the pipe network; s6: the method comprises the steps of executing a cyclic execution step of dosing operation by a pipe network, establishing a pipe network water quality intelligent monitoring, regulating and controlling response mechanism based on multivariable model predictive control, designing and realizing a system based on multivariable fusion dosing model control, constructing an intelligent dosing model and a table according to a large amount of historical data acquired by a plurality of sensors arranged on pipe network water quality monitoring sites, inquiring the table according to dynamic real-time monitoring data to determine dosing weight, performing feedback optimization on the model according to the real-time data, and having a function of predicting the dosing weight of a heat supply pipe network according to the model, thereby achieving the aim of accurately controlling the pipe network water quality.
Description
Technical Field
The invention belongs to the technical field of intelligent water quality monitoring and control, and particularly relates to an intelligent heat supply pipe network water quality monitoring and control system and method.
Background
The heating system pipe network equipment comprises water treatment equipment, a heat exchanger, valves of various sizes, various control instruments, a control system computer, a large display screen, large and small pipelines and a water pump. The heating equipment is operated safely, reliably, economically and reasonably, the environment is protected, and the pollution is reduced. Needs to select an economic and environment-friendly water treatment method to perform the work of scale prevention, corrosion prevention and artificial water loss prevention. The heating system can supply heat with high quality in a non-scale, non-rust and non-corrosion state, thereby achieving the comprehensive benefits of saving energy, saving water, reducing emission, reducing maintenance amount and prolonging the service life of equipment. Therefore, the water quality needs to be dynamically monitored in real time during the operation of a pipe network, and proper agents are added according to the water quality condition, so that the drugs can slowly react in the pipeline to play a role in rust prevention and scale inhibition, thereby achieving the purpose of regulating and controlling the water quality.
At present, the main monitoring and dosing mode is that a medicament with fixed weight is manually added during water replenishing according to experience, the mode is simple and convenient to operate, but has poor effect, cannot respond in time according to the change of water quality, and has unstable PH value and calcium ion value in a pipeline, so that the pipeline scaling is easy to cause to influence the heat supply effect, and the service life of a heat supply pipeline is shortened; the reagent adding is unstable, the detection effect is lagged, the annual reagent consumption cost of the large-scale pipe network heat exchange station is high, and the cost control is not facilitated by the judgment of the adding amount through human experience.
With the development of modern informatization and automation technology, particularly the development of computers, internet of things, sensors, instruments and meters and automatic control technology, the new generation information technology is widely and deeply applied to the fields of industrial production, automatic control, equipment management, intelligent detection, intelligent monitoring, smart cities and the like, and brand new technical means and practical tools are brought to the automatic control and intelligent management of a heating system. The intelligent management system has important practical significance for improving the automation, informatization and intellectualization levels of the management of the large-scale heating system, and further has indirect promotion significance for the energy-saving and environment-friendly career.
In summary, aiming at the problems of water quality monitoring and regulation of pipe networks, the prior art and the prior art still have great defects in the aspects of dynamic real-time continuous detection, an automatic timely response mechanism, an accurate regulation strategy, cost control, environmental protection requirements and the like, so the invention designs and discloses a method and a system for intelligent water quality monitoring and control of a pipe network, which comprehensively utilize a computer, an internet of things, sensors, instruments and meters and an automatic control technology, establish a response mechanism for intelligent water quality monitoring and control of the pipe network based on model prediction control, design and realize a system for intelligent dosing control of the pipe network based on multivariable monitoring data, realize dynamic regulation and control of dosing weight according to dynamic real-time monitoring data collected by a plurality of sensors arranged at water quality monitoring sites of the pipe network, and realize a dosing mode of the pipe network with a pre-regulation function according to model prediction, thereby realizing the aims of automatic, informatization, intelligent real-time water quality monitoring of the pipe network and dynamic accurate control.
Disclosure of Invention
The invention aims to solve the problems of water quality monitoring and dynamic accurate regulation and control of a pipe network, provides an intelligent monitoring and control method and system for the water quality of the pipe network, aims at the requirements of real-time water quality monitoring and dynamic accurate regulation and control of the pipe network, designs automatic, digital, intelligent and convenient multi-class water quality index monitoring and dynamic model regulation and control method flows for realizing timeliness, accuracy and predictive effect of the monitoring and regulation and control process, and builds an intelligent monitoring and control system for the water quality of the pipe network to realize scientific and reasonable water quality monitoring and dynamic accurate regulation and control targets of the pipe network. And a brand new technical means and a practical tool are provided for the design, optimization and management of the heating system. Under the big situation of heating system intelligent upgrading transformation, intelligent medicine becomes the heat supply link and heavily weighs in, and for further promoting quality of water, the accurate dosing system who has model predictive control algorithm will be adopted to the core medicine technology, helps promoting the accurate regulation and control effect of large-scale pipe network and management efficiency, reduces the cost and the unknown risk of pipe network operation management, has important realistic meaning to energy-concerving and environment-protective career and wisdom city construction.
The invention is realized by the following technical scheme, and provides a heat supply pipe network water quality real-time monitoring and dynamic accurate control method, which comprises a heat supply pipe network water quality intelligent monitoring control system implementing the method, and a core dosing model and a corresponding table forming the system.
The flow of the steps of the pipe network water quality real-time monitoring and dynamic accurate control method is shown in figure 1, and the method specifically comprises the following steps:
s1: monitoring the water quality of the heat supply pipe network;
s2: uploading water quality monitoring data of a pipe network;
s3: judging the water quality abnormity of the pipe network;
s4: constructing and optimizing a dosing model;
s5: automatically checking a table by a program to determine the dosing weight of the pipe network;
s6: the pipe network executes the dosing operation.
In the operation process of the heating system, the steps of the pipe network water quality real-time monitoring and dynamic accurate control flow are continuously and circularly operated, namely after the steps S1 to S6 are finished, the step S1 is returned to, the steps S1 to S6 are continuously executed, and the circulation is continuously operated.
The real-time monitoring and dynamic accurate control method for the water quality of the pipe network is realized based on an intelligent monitoring and control system for the water quality of the pipe network, and as shown in figure 2, the real-time monitoring and dynamic accurate control method for the water quality of the pipe network comprises the following components: 1 medicament holding vessel, 2 add medicine controller, 3 add medicine pump, 4 pipe network internal piping, 5 circulating pumps, 6 return water end sensor groups, 7 controller data lines, 8 moisturizing pumps, 9 external data lines of server, 10 heat exchange station servers, 11 controllers, 12 sensor data lines, 13 external moisturizing pipes, 14 heat exchange stations.
The invention provides a real-time monitoring and dynamic accurate control method for water quality of a pipe network, which comprises the following steps of S1: heat supply pipe network water quality monitoring utilizes the 6 return water end sensor groups of installing on 4 pipe network internal pipe ways to carry out water quality monitoring, and sensor group includes: circulating pump power, temperature sensor, pressure sensor, flow sensor, PH value sensor, calcium ion (Ca 2 +) concentration sensor. Wherein, the temperature sensor and the pressure sensor are arranged at each 1 measuring point on the return water main pipeline; the flow sensor is arranged on the water supply pipeline; the PH value and the concentration of calcium ions (Ca & lt 2+ & gt) are directly led out of the opening on the pipeline to the measuring equipment.
The step S2: the uploading of the water quality monitoring data of the pipe network is realized by transmitting various data to the in-station controller after the data are measured by the sensors and uploading the sensing data to the server.
The step S3: the method comprises the steps of judging the water quality abnormity of the pipe network, wherein the reason of water quality mutation of the pipe network is influenced by various factors, monitoring a water medium of the pipe network through a pH value sensor and a Ca2+ ion concentration sensor, judging whether sudden change occurs, generally detecting that a water quality index changes in 1 cycle period, judging that the water quality index changes suddenly when the change value of the pH value or the Ca2+ ion concentration exceeds a judgment threshold value V in a measurement time T, and triggering the water quality abnormity of the pipe network.
The step S4: the dosing model construction and optimization comprises 2 sub-steps, namely S4.1 realizes the dosing model construction based on multivariable historical monitoring data, and S4.2 uses real-time monitoring data to optimize and adjust the dosing model.
S4.1, construction of an intelligent dosing model based on historical monitoring data is achieved, namely the intelligent dosing model is a nonlinear model based on multivariable monitoring data, aiming at the characteristics of multivariable and large hysteresis of a dosing control link of a heat supply pipe network, an intelligent dosing system based on multivariable model predictive control is provided, and application parameter configuration of the whole control system is set aiming at dosing control indexes of the heat supply pipe network.
The model matrix parameters of the model predictive control are mainly as follows: controlled Variables (CVs), arguments (MVs), disturbance Variables (DVs).
Indirect variables: pipeline diameter, pipeline length, and circulating pump power;
controlled Variables (CVs): the pH value of the water and the concentration of Ca < 2+ > ions;
independent variables (MVs): adding medicine weight;
disturbance Variables (DVs): pipe network temperature, pressure, flow. Disturbance variables are also uploaded through data acquisition of various instruments and monitoring sensors.
Controlled Variables (CVs) refer to process variables that need to be held at a target or within a set range; independent variables (MVs) refer to measured process variables that can be adjusted to affect Controlled Variables (CVs); disturbance Variables (DVs) refer to measured process variables that affect CVs that cannot be adjusted.
The system uses a process dynamics model to predict the behavior of the process outputs (CVs) and find the best (MVs), driving the CVs to a target value.
S4.1, the dosing model construction based on the multivariable historical monitoring data is realized by utilizing a long-term actual monitoring process, a large amount of historical monitoring data of indexes such as temperature, pressure, flow, PH value, ca2+ ion concentration and the like of a pipe network medium are utilized, the historical water quality monitoring data is used as model input, the corresponding dosing weight is used as model output, the data are subjected to standardization processing, a plurality of variables such as pressure, temperature and flow are considered through cluster analysis and regression analysis, the optimal action interval division suitable for the multivariable and the corresponding optimal dosing weight are determined, a nonlinear dosing weight query table is constructed, transverse table heads of the query table correspond to the plurality of variables to be considered, and the longitudinally distributed records represent multivariable group value intervals and the corresponding dosing weight.
S4.2, the dosing model is optimized and adjusted by using real-time monitoring data, namely, according to the change of the current water quality monitoring data result generated by a recently completed dosing operation on the water quality monitoring data before dosing, if the regulation and control effect is judged to be superior to the set regulation and control threshold value, all water quality indexes monitored by the current sensor group are used as input, and according to the same data standardization processing mode adopted during model construction, the record is inserted into a multivariable dosing weight query table, so that the dosing weight corresponding to the currently monitored parameter level is refined.
Preferably, in the step 4.2, the dosing model is optimized and adjusted by using the real-time monitoring data, an operation time interval can be set according to actual needs, the model construction operation is updated, and the updated monitoring data is added into the multivariable dosing weight lookup table, so that a new model is generated, and the purpose of model optimization is achieved.
The intelligent dosing model has model prediction capability, and a model prediction control algorithm is generally divided into the following three parts:
(1) A prediction model: all the predictive control algorithms need to obtain a predictive model in advance, in the embodiment, a dosing model of long-term historical data aiming at a plurality of variable factors such as temperature, pressure, flow and the like is used, a data table is built, and the model automatically looks up the table according to the current monitoring data by a program deployed on a server to obtain the predictive output.
(2) And (3) rolling optimization: that is, at each sampling time point, the algorithm only optimizes a limited time domain within a range, after all optimization controls are obtained, only the current control quantity is implemented in the system, and optimization prediction is carried out again at the next sampling time point.
(3) And (3) feedback correction: and calculating the output of the next sampling time point according to the prediction model, and solving the error between the actually measured output and the predicted output at the next sampling time point so as to perform heuristic correction on the future output.
The step S5: and (4) automatically checking a table by a program to determine the dosing weight of the pipe network, inquiring a multivariable dosing weight inquiry table by using the multivariable intelligent dosing model constructed in the step (S4) according to the currently monitored pipe network state data including the controlled variable and the disturbance variable, and searching the dependent interval of each variable of the currently monitored data to obtain the dosing weight of the pipe network recorded in the interval in the table.
The step S6: and C, the pipe network executes the drug adding operation, the determined drug adding weight is automatically checked by the program in the step S5, the controller drives the pipe network pump set to execute the drug adding operation, and the controller drives the 3 drug adding pumps to start and stop according to the calculated time.
And the step S1 to the step S6 are executed once to form a work period cycle, and the system is in a cycle work mode of real-time monitoring.
Compared with the prior art, the invention has the beneficial effects that: the invention applies big data and nonlinear model technology, and aims at the characteristics of multivariable and big hysteresis of the dosing control link of the pipe network, an advanced intelligent dosing system based on multivariable model predictive control technology is provided, the application parameter configuration of the whole control system is customized and developed for dosing of the heat supply pipe network, and the model predictive control of the invention has the following typical characteristics:
(1) A predictive algorithm that can obtain a non-parametric model based on a simple pulse or step signal, without the need to further identify the model, such a signal being easy to implement for industrial processes;
(2) The idea of multi-step optimization in predictive control is adopted, so that the robustness of the controller is enhanced;
(3) The optimal control algorithm is realized by a computer, meanwhile, the optimal control calculation is real-time on-line calculation, all control parameters and monitoring parameters are dynamically acquired in real time, and the control algorithm is multi-input multi-output feedforward control;
(4) The modeling is convenient, and the internal mechanism of the process does not need to be deeply understood;
(5) The nonlinear model is constructed in a form of a multivariable table, so that the robustness of the system is improved;
(6) A rolling optimization strategy has a good dynamic control effect, and a dynamic model directly considers the process dynamics;
(7) The simple and practical model correction method has strong robustness, and the set value is continuously optimized, calculated and adjusted according to the model;
(8) The method can be applied to the processes of band constraint, large pure lag, non-minimum phase, multiple input multiple output, nonlinearity and the like.
The multivariable model based predictive control can be used for controlling a plurality of variables of the whole controlled object, the mutual influence among a plurality of loops is eliminated, the predictive control has the predictive capability, and the predictive control can be analyzed according to the working conditions of the plurality of loops at present, so that the future trend of each loop in the controller is predicted, the loops are adjusted according to the predicted result, and compared with the traditional program adjustment, the multivariable predictive control has stronger adaptability and better robustness, is suitable for the large lag and strong coupling characteristics of the processing process, can effectively solve the process measurable interference, adopts a multivariable optimization algorithm, is suitable for processing the problems of multi-level, multi-target and multi-constraint control, and can more conveniently enable the production process control to reflect the economic indexes of the production process.
The invention provides a brand new technical means and a practical tool for the design, optimization and management of a dosing scheme of a pipe network of a large-scale heating system. The system has important practical significance for improving the dosing effect and the utilization efficiency of a large-scale heating system pipe network and reducing the production operation cost and the sudden risk. The invention promotes the wide and deep application of new generation information technologies such as big data artificial intelligence and the like in the fields of intelligent management of heating systems and even smart cities.
Drawings
FIG. 1 is a flow chart of the steps of the method of the present invention;
FIG. 2 is a schematic view of an intelligent monitoring and controlling system for water quality of a heat supply pipe network;
FIG. 3 is a pH curve of water quality controlled manually in practical application;
FIG. 4 is a diagram of a pH curve of water quality predicted and controlled by a model considering multivariate factors according to the present invention in practical application;
FIG. 5 is a curve of Ca2+ ion concentration value of water quality manually controlled in practical application;
FIG. 6 is a curve of the concentration value of Ca2+ ions in water predicted and controlled by the model considering multivariate factors in practical application.
Detailed Description
The invention is described in further detail below with reference to the accompanying drawings:
the invention discloses a heat supply pipe network water quality intelligent monitoring control system and a method, which comprehensively utilize a computer, an Internet of things, sensors, instruments and meters and an automatic control technology, establish a pipe network water quality intelligent monitoring control response mechanism based on model predictive control, design a matched pipe network dosing control device system based on multivariate monitoring, and realize the dynamic regulation and control of the proportion and the adding mode of various different medicaments according to dynamic real-time monitoring data and historical data acquired by various sensors arranged at heat supply pipe network water quality monitoring sites.
The specific implementation process method of the method is shown in figure 1, wherein the system of the intelligent dosing model pipe network dosing control device based on multivariate fusion is shown in figure 2.
In the specific implementation process, the length of a pipe network is 100-4000 meters, the pipe network generally only has 1-level pipeline, the diameter of the pipe network is DN80-DN350, generally 1 circulating pump, 1 water replenishing pump and 1 dosing pump can be arranged in one loop, 1 dosing point is arranged in one loop, and the dosing pump is controlled by a controller in a station to add the medicament into the pipeline.
The step S2: the method comprises the steps that pipe network water quality monitoring data are uploaded, various data are measured by a water supply end sensor and a water return end sensor and then transmitted to an in-station controller, meanwhile, sensing data are uploaded to a server and uploaded to a smart platform in real time, the data collected by the sensors are analog signals, and a Modbus protocol is adopted for uploading the data of the server; the remote data can be uploaded by using GPRS communication protocol.
The medicament can be prepared from common additive medicaments, such as: anticorrosive antisludging agent for circulating water.
In general, taking a secondary pipe network with a distance of 1000-4000 meters as an example, the following steps are taken:
dosing frequency: 100-150g of water per ton;
detecting items: PH, ca2+ ion concentration;
monitoring frequency: the monitoring frequency of the system is once every 30 seconds, the weighted average value is taken every 1 hour, the execution period of the pipe network dosing is executed every 1 hour, and the dosing action is executed when the water quality is found to be abnormal.
Under the large situation of intelligent upgrading and transformation of a heat supply system, intelligent dosing of a pipe network becomes the middle of the heat supply link, and in order to further improve water quality, protect the environment and save cost, a core dosing process adopts an accurate dosing system with a model prediction control algorithm. The intelligent management monitoring and dynamic regulation targets are as follows:
(1) By utilizing big data cluster analysis and regression analysis algorithm, a model is established to automatically determine the type and weight of the added medicament by comprehensively judging factors such as circulating water temperature, circulating water pressure, circulating water flow, PH value, ca < 2+ > ion concentration and the like of the pipe network.
(2) Finally, a scientific calculation basis in principle is formed, quantitative big data analysis is supported, the economic benefit of future production operation is quantitatively evaluated after the system is used, and the system can assist in organizing and summarizing innovative applications.
The intelligent monitoring and control system and method for water quality of the pipe network support the intelligent dosing expected benefit analysis of enterprises, and the final implementation effect achieves the following aims:
(1) Index control: the pH value and the Ca < 2+ > ion concentration are stable and reach the standard, and the condition of large-amplitude fluctuation is reduced. The volatility index is compared with real statistical data before and after implementation by taking days as a standard, and monthly water quality and drug consumption data are statistically analyzed;
(2) The water quality stability of the pipe network, the dosage-saving index and the model adaptability are continuously optimized through the model parameters, the water quality index is visually displayed through the historical trend, the medicine consumption is reduced by more than 10 percent compared with the traditional feeding mode, and the calculation is carried out again according to the real production data under the specific implementation condition.
(3) Realize the full-automatic closed-loop adjustment of medicine system, reduce personnel's intervention.
In the intelligent monitoring and controlling method and system for the water quality of the pipe network, the system can realize the following functions:
(1) The system can give an early warning according to the liquid medicine storage quantity of the medicine storage tank and realize the medicine feeding control.
(2) The system can realize automatic dispensing according to the stock state of the medicine storage tank and the production requirement.
(3) The system can automatically control the start, stop and switch of the feeding pump according to the state and the feeding amount of the feeding pump group.
(4) The system can realize the switching of the main and standby adding pipelines and the switching of the adding system according to the state of the pipelines.
(5) The system can automatically calculate the adding amount according to the change of water quantity and water quality under the normal working condition, and implements adding of each medicine adding pipeline according to the adding amount, so that the water quality is automatically adjusted, and the water quality of a pipe network is ensured to meet the requirement.
(6) The system reduces the consumption of added drugs on the premise of ensuring that the water quality of the pipe network reaches the standard.
In the intelligent dosing model based on the multivariable fusion, in the specific implementation process, the model matrix parameters of the model predictive control are mainly as follows: controlled Variables (CVs), arguments (MVs), disturbance Variables (DVs).
Controlled Variables (CVs): PH, ca2+ ion concentration;
independent variables (MVs): adding medicine weight;
disturbance Variables (DVs): the return water temperature, the return water pressure and the flow of the pipe network. Disturbance variables are also uploaded through data acquisition of various instruments and monitoring sensors.
Controlled Variables (CVs) refer to process variables that need to be held at a target or within a set range; independent variables (MVs) refer to measured process variables that can be adjusted to affect CVs; disturbance Variables (DVs) refer to measured process variables that affect CVs that cannot be adjusted.
Model predictive control uses a process dynamics model to predict the future behavior of process outputs (CVs) and find the best independent variables (MVs) to drive the CVs to target values while considering constraints.
In this embodiment, S4.1 implements a dosing model construction based on multivariate historical monitoring data, an intelligent dosing model is constructed using annual heating monitoring data, in an actual monitoring process, indexes such as a return water temperature, a return water pressure, a flow rate, a heat value, a PH value, and a Ca2+ ion concentration of a pipe network are monitored, and uploaded to a server, historical water quality monitoring data is used as a model input according to a large amount of historical monitoring data, a corresponding dosing amount is used as a model output, data is standardized, sample data is subjected to mean value and distribution variance calculation according to data dimension units, a cluster analysis and regression analysis are performed to consider a plurality of variables such as a return water pressure, a return water temperature, a flow rate, and a total water amount, on the basis of a dosing reference amount, an optimal action interval division applicable to multivariate and a corresponding optimal dosing weight are determined, a nonlinear dosing multivariable weight query table is constructed, a transverse table head of the query table corresponds to a plurality of variables and adjustment values to be considered, a longitudinal distribution record represents a multivariable group value interval and a corresponding adjustment value and a weight coefficient, and a number of records of a plurality of different longitudinal combination parameters. And calculating the corresponding dosing weight according to the reference amount, the adjustment value and the coefficient. The look-up table is deployed on the server.
In this embodiment, multiple dosing categories correspond to multiple regulatory targets, one dosing parameter table for each category of drug, table 1 for PH regulation, and table 2 for Ca2+ ion concentration (i.e., hardness) regulation.
In this embodiment, a parameter lookup table shown in table 1 is adopted, and the horizontal header includes a PH value, a dosing reference amount 1 (g), a return water temperature (degree), an adjustment value 1_1, a return water pressure (mPa), an adjustment value 1_2, a flow rate (t/h), an adjustment value 1_3, a total water amount (ton), and a coefficient 1, where the adjustment value 1_1 is determined according to the return water temperature, the adjustment value 1 u 2 is determined according to the return water pressure, the adjustment value 1_3 is determined according to the flow rate, and the coefficient 1 is determined according to the total water amount.
TABLE 1 multivariable medicine-feeding parameter table of PH value
In this embodiment, a parameter lookup table shown in table 2 is adopted, and the horizontal header includes a Ca2+ ion concentration value, a dosing reference amount 2 (g), a return water temperature (degree), an adjustment value 2_1, a return water pressure (mPa), an adjustment value 2_2, a flow rate (t/h), an adjustment value 2_3, a total water amount (ton), and a coefficient 2, where the adjustment value 2_1 is determined according to the return water temperature, the adjustment value 2_2 is determined according to the return water pressure, the adjustment value 2_3 is determined according to the flow rate, and the coefficient 2 is determined according to the total water amount.
TABLE 2 multivariable dosing parameter table of Ca2+ ion concentration value
Table 1 and table 2 determine:
the dose weight 1= (reference amount 1 + adjustment value 1 u 2+ adjustment value 1 u 3) × 1.
The dosed weight 2= (reference amount 2+ adjustment value 2 u 1 + adjustment value 2 u 2+ adjustment value 2 u 3) × 2.
Total dosing weight = dosing weight 1 + dosing weight 2.
In the embodiment, a K-means clustering mode is adopted to determine distribution centers and interval ranges of multiple variables such as return water pressure, return water temperature, flow and total water quantity, 10-20 data intervals are set according to actual data calculation and analysis, a multiple nonlinear regression analysis mode is adopted to determine weighting coefficients among the factors, a dosing data table is constructed, and the table head comprises multiple disturbance variables such as return water pressure, return water temperature and flow, controlled variables such as water quality PH value and water quality Ca2+ ion concentration and dosing weight. The records of longitudinal distribution represent multivariable group value intervals and corresponding dosing weights, and the total record number of the table is calculated according to the interval distribution number of each variable factor.
S4.2, the dosing model is optimized and adjusted by using real-time monitoring data, namely, according to the change of the current water quality monitoring data result generated by a recently completed dosing operation on the water quality monitoring data before dosing, if the regulation and control effect is superior to the set regulation and control threshold value, all water quality indexes monitored by the current sensor group are taken as input, according to the same data standardization processing mode adopted during model construction, the record is inserted into a multivariable dosing weight query table, and the clustering center and interval boundary are recalculated, so that the dosing weight corresponding to the currently monitored parameter level is refined.
Preferably, in the step 4.2, the dosing model is optimized and adjusted by using the real-time monitoring data, an operation time interval can be set according to actual needs, the model construction operation is updated, and the updated monitoring data is added into the multivariable dosing weight lookup table, so that a new model is generated, and the purpose of model optimization is achieved.
The step S5: and (4) automatically checking a table by a program to determine the dosing weight of the pipe network, inquiring a multivariable dosing weight inquiry table by using the multivariable intelligent dosing model constructed in the step (S4) according to the currently monitored pipe network state data including the controlled variable and the disturbance variable, retrieving the dependent interval of each variable of the currently monitored data, obtaining the dosing weight of the pipe network recorded in the interval in the table, deploying the table checking program on a server, automatically finishing the table checking process by the program, and driving a controller to execute operation by the program according to the result.
In this embodiment, as shown in fig. 3, a PH curve of water quality manually controlled in practical application is shown; FIG. 4 is a diagram of a model predictive control water pH curve in practical application taking multivariable factors into consideration; FIG. 5 is a curve of concentration values of Ca2+ ions in water manually controlled in practical application; FIG. 6 is a curve of Ca2+ ion concentration value for controlling water quality by using multivariate factor model prediction in practical application. The monitoring data result in the graph shows that after the method and the system are used, the pH value curve of the water quality and the Ca < 2+ > ion concentration value curve of the water quality are subjected to predictive control by an intelligent dosing model based on multivariate factors, compared with the manual control effect, the regulation and control effect is obviously improved, and the method and the system have good practicability and effectiveness.
The intelligent monitoring and controlling system for the water quality of the pipe network has a periodic operation cost accounting function, and carries out cost accounting according to various cost expenditures consumed in a specified time period, wherein the cost accounting comprises the cost of adding medicaments, water charges, electricity charges and the like.
The intelligent monitoring and controlling system for the water quality of the pipe network has a periodic operation prediction cost accounting function, and carries out cost accounting according to various cost expenses which are predicted to be generated by using a model in a specified time period, wherein the cost accounting comprises the cost of adding medicaments, water charges, electricity charges and the like. And auxiliary decision basis is provided for macro management.
In conclusion, the invention realizes an intelligent heat supply pipe network water quality monitoring and controlling system and a method, provides a pipe network water quality real-time monitoring and dynamic accurate controlling method, and comprises a pipe network water quality intelligent monitoring and controlling system for implementing the method. The intelligent monitoring and control response mechanism of the water quality of the pipe network based on model predictive control is established by comprehensively utilizing a computer, the Internet of things, sensors, instruments and meters and an automatic control technology, a system of the intelligent dosing control device of the pipe network based on multivariate fusion is designed and realized, an intelligent dosing model and a data table of the multivariate fusion are constructed according to historical data acquired by various sensors arranged at water quality monitoring sites of the pipe network, a dosing strategy is determined according to a dynamic real-time monitoring data program, feedback optimization is carried out on the model, dynamic regulation and control of the proportion and the dosing mode of various different medicaments are realized, and the dosing mode of the pipe network with a pre-regulation and control function is realized according to model prediction, so that the aims of automation, informatization, intelligent real-time monitoring of the water quality of the pipe network and dynamic accurate control are realized.
The above-described embodiment is only one embodiment of the present invention, and it will be apparent to those skilled in the art that various modifications and variations can be easily made based on the application and principle of the present invention disclosed in the present application, and the present invention is not limited to the method described in the above-described embodiment of the present invention, so that the above-described embodiment is only preferred, and not restrictive.
Claims (8)
1. An intelligent monitoring and control system and method for water quality of a heat supply pipe network are characterized in that: the method comprises a heat supply pipe network water quality intelligent monitoring control system for implementing the method, and the flow of the pipe network water quality real-time monitoring and dynamic accurate control method comprises the following steps: s1: heat supply pipe network water quality monitoring, S2: uploading water quality monitoring data of a pipe network, and S3: judging the water quality abnormity of the pipe network, and S4: constructing and optimizing a dosing model, S5: automatically checking the table by a program to determine the dosing weight of the pipe network, and S6: the pipe network executes the dosing operation; the steps of the pipe network water quality real-time monitoring and dynamic accurate control flow are in a continuous cycle operation working mode, the steps from S1 to S6 are executed once to be a working cycle, namely after the steps from S1 to S6 are finished, the step returns to S1, the steps from S1 to S6 are continuously executed, and the cycle is continuously operated.
2. The intelligent heat supply pipe network water quality monitoring and control system and method according to claim 1, characterized in that: and S1, monitoring the water quality of the heat supply pipe network by using a sensor.
3. The intelligent monitoring and control method and system for the water quality of the heat supply pipe network according to claim 1 are characterized in that: and S2, uploading the water quality monitoring data of the pipe network, namely, transmitting the data to the in-station controller after the data are measured by the water return end sensor, and uploading the data to the management server.
4. The intelligent heat supply pipe network water quality monitoring and control system and method according to claim 1, characterized in that: and S4, constructing and optimizing a dosing model, wherein the step comprises 2 sub-steps, namely S4.1 realizes the construction of the dosing model based on multivariate historical monitoring data, and S4.2 optimizes and adjusts the dosing model by using real-time monitoring data.
5. The S4.1 of claim 4, which is used for implementing the dosing model construction based on the multivariate historical monitoring data, means that the intelligent dosing model is a nonlinear model based on the multivariate monitoring data; aiming at the characteristics of multivariable and large hysteresis of a dosing control link of a heat supply pipe network, an intelligent dosing system based on multivariable model predictive control is provided, and the application parameter configuration of the whole control system is set aiming at the dosing control index of the heat supply pipe network; by utilizing a long-term actual monitoring process, a large amount of historical monitoring data of indexes such as temperature, pressure, flow, PH value, ca2+ ion concentration and the like of a pipe network medium are input as a model, corresponding dosing weight is output as a model, the data are subjected to standardization processing, a plurality of variables such as pressure, temperature and flow are considered, a multivariate clustering center and optimal action interval division are determined, further, corresponding optimal dosing weight is determined, a nonlinear multivariable weight query table is constructed, a transverse table head of the query table corresponds to a plurality of variables to be considered and corresponding regulating values, and a longitudinally distributed record represents a multivariable group interval and corresponding dosing weight.
6. The S4.2 of claim 4, optimally adjusting the dosing model using the real-time monitoring data, that is, according to the change of the current water quality monitoring data result generated by a recently completed dosing operation on the water quality monitoring data before dosing, if the regulation and control effect is better than the set regulation and control threshold, taking each water quality index monitored by the current sensor group as input, and inserting the record into a multi-variable dosing amount lookup table according to the same data standardization processing mode adopted during model construction, thereby refining the corresponding dosing weight at the currently monitored parameter level.
7. The intelligent heat supply pipe network water quality monitoring and control system and method according to claim 1, characterized in that: the step S5: and (5) automatically checking a table by a program to determine the dosing weight of the pipe network, using the multivariable intelligent dosing model constructed in the step (S4), inquiring a multivariable dosing weight inquiry table according to the currently monitored pipe network state data including the controlled variable and the disturbance variable, retrieving the dependent interval of each variable of the currently monitored data, and obtaining the dosing weight of the pipe network recorded in the interval in the table.
8. The intelligent heat supply pipe network water quality monitoring and control system and method according to claim 1, characterized in that: the step S6: and (5) the pipe network pump set executes the drug adding operation, and the controller drives the pipe network pump set to execute the drug adding operation according to the drug adding weight determined in the step S5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211221992.1A CN115579072A (en) | 2022-10-08 | 2022-10-08 | Intelligent water quality monitoring control system and method for heat supply pipe network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211221992.1A CN115579072A (en) | 2022-10-08 | 2022-10-08 | Intelligent water quality monitoring control system and method for heat supply pipe network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115579072A true CN115579072A (en) | 2023-01-06 |
Family
ID=84582619
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211221992.1A Pending CN115579072A (en) | 2022-10-08 | 2022-10-08 | Intelligent water quality monitoring control system and method for heat supply pipe network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115579072A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116414420A (en) * | 2023-06-09 | 2023-07-11 | 山东华邦农牧机械股份有限公司 | Automatic upgrading method of poultry breeding control system |
-
2022
- 2022-10-08 CN CN202211221992.1A patent/CN115579072A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116414420A (en) * | 2023-06-09 | 2023-07-11 | 山东华邦农牧机械股份有限公司 | Automatic upgrading method of poultry breeding control system |
| CN116414420B (en) * | 2023-06-09 | 2023-10-13 | 山东华邦农牧机械股份有限公司 | Automatic upgrading method of poultry breeding control system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109521813B (en) | System and method for stably controlling concentration ratio of circulating water of thermal power plant | |
| CN109062053B (en) | Denitration ammonia injection control method based on multivariate correction | |
| CN113657031A (en) | Digital twin-based heat supply scheduling automation realization method, system and platform | |
| CN112128841A (en) | Whole-network balance adjusting method based on load prediction and room temperature feedback correction | |
| CN102444784B (en) | Pressure control system for steel enterprise steam pipe network based on dynamic matrix control | |
| CN100545772C (en) | A kind of coal-burning boiler system mixing control method | |
| CN105303243B (en) | Method for predicting balance scheduling system by using blast furnace gas | |
| CN112232980A (en) | Regulation and control method for heat pump unit of regional energy heat supply system | |
| CN115013861B (en) | Indoor temperature control method and device based on heating system | |
| CN109886471A (en) | Fired power generating unit load distribution method based on neural network and intelligent optimization algorithm | |
| CN102540879A (en) | Multi-target evaluation optimization method based on group decision making retrieval strategy | |
| CN101709863B (en) | Hybrid control method for furnace pressure system of coal-fired boiler | |
| CN105988378A (en) | Circulating cooling water control system and drug adding control device and drug adding control method thereof | |
| CN115579072A (en) | Intelligent water quality monitoring control system and method for heat supply pipe network | |
| CN115207977A (en) | Active power distribution network deep reinforcement learning real-time scheduling method and system | |
| CN113606634A (en) | Heat source guarantee system based on long-distance transmission pipe network optimization heat source operation parameter | |
| CN106200379A (en) | A kind of distributed dynamic matrix majorization method of Nonself-regulating plant | |
| CN104408317A (en) | Metallurgy enterprise gas flow interval predicting method based on Bootstrap echo state network integration | |
| Sun et al. | Integrated control strategy of district heating system based on load forecasting and indoor temperature measurement | |
| CN108734419B (en) | Blast furnace gas scheduling system modeling method based on knowledge migration | |
| CN112783115A (en) | Online real-time optimization method and device for steam power system | |
| CN101819409A (en) | Chlorination control method and device | |
| CN113685971A (en) | Constant-temperature heat supply automatic control circulation dynamic balance control system and method and data processing terminal | |
| CN113112095A (en) | Multi-mode multi-heat supply unit load real-time optimization scheduling method and management system | |
| CN114779722B (en) | Intelligent combustion optimization control system and method for coal-fired power station boiler |
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 |