CN104697107A - Intelligent learning energy-saving regulation and control system and method - Google Patents
Intelligent learning energy-saving regulation and control system and method Download PDFInfo
- Publication number
- CN104697107A CN104697107A CN201310724423.3A CN201310724423A CN104697107A CN 104697107 A CN104697107 A CN 104697107A CN 201310724423 A CN201310724423 A CN 201310724423A CN 104697107 A CN104697107 A CN 104697107A
- Authority
- CN
- China
- Prior art keywords
- data
- adaptive
- energy
- air
- conditioning system
- 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.)
- Granted
Links
- 230000013016 learning Effects 0.000 title claims abstract description 72
- 238000000034 method Methods 0.000 title claims abstract description 69
- 230000003044 adaptive effect Effects 0.000 claims abstract description 100
- 238000004378 air conditioning Methods 0.000 claims abstract description 63
- 238000009795 derivation Methods 0.000 claims abstract description 20
- 238000005457 optimization Methods 0.000 claims abstract description 6
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 55
- 238000005086 pumping Methods 0.000 claims description 26
- 239000000498 cooling water Substances 0.000 claims description 23
- 239000000284 extract Substances 0.000 claims description 18
- 239000005457 ice water Substances 0.000 claims description 17
- 238000005516 engineering process Methods 0.000 claims description 16
- 230000009467 reduction Effects 0.000 claims description 6
- 238000004134 energy conservation Methods 0.000 claims description 5
- 230000001105 regulatory effect Effects 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 4
- 238000012731 temporal analysis Methods 0.000 claims 1
- 238000000700 time series analysis Methods 0.000 claims 1
- 238000005265 energy consumption Methods 0.000 abstract description 3
- 238000004806 packaging method and process Methods 0.000 abstract 1
- 230000008569 process Effects 0.000 description 11
- 238000004422 calculation algorithm Methods 0.000 description 10
- 238000004364 calculation method Methods 0.000 description 5
- JEGUKCSWCFPDGT-UHFFFAOYSA-N h2o hydrate Chemical compound O.O JEGUKCSWCFPDGT-UHFFFAOYSA-N 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000001276 controlling effect Effects 0.000 description 2
- 238000010219 correlation analysis Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000002708 enhancing effect Effects 0.000 description 2
- 230000008676 import Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000004087 circulation Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000001066 destructive effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000012367 process mapping Methods 0.000 description 1
- 238000004064 recycling Methods 0.000 description 1
- 239000003507 refrigerant Substances 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/70—Control systems characterised by their outputs; Constructional details thereof
- F24F11/80—Control systems characterised by their outputs; Constructional details thereof for controlling the temperature of the supplied air
- F24F11/83—Control systems characterised by their outputs; Constructional details thereof for controlling the temperature of the supplied air by controlling the supply of heat-exchange fluids to heat-exchangers
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/30—Control or safety arrangements for purposes related to the operation of the system, e.g. for safety or monitoring
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F2110/00—Control inputs relating to air properties
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
- F24F11/63—Electronic processing
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/70—Control systems characterised by their outputs; Constructional details thereof
- F24F11/80—Control systems characterised by their outputs; Constructional details thereof for controlling the temperature of the supplied air
- F24F11/83—Control systems characterised by their outputs; Constructional details thereof for controlling the temperature of the supplied air by controlling the supply of heat-exchange fluids to heat-exchangers
- F24F11/85—Control systems characterised by their outputs; Constructional details thereof for controlling the temperature of the supplied air by controlling the supply of heat-exchange fluids to heat-exchangers using variable-flow pumps
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Signal Processing (AREA)
- Life Sciences & Earth Sciences (AREA)
- Atmospheric Sciences (AREA)
- Air Conditioning Control Device (AREA)
Abstract
The invention provides an intelligent learning energy-saving regulation and control system and method. In the method, firstly, a system regulation and control data setting module is used for setting system regulation and control data and extracting real-time sensing data from an air conditioning system; then enabling an experience learning and logic derivation module to carry out experience learning and logic derivation according to the system regulation data and the real-time sensing data, and establishing and packaging a temperature difference and energy consumption adaptability derivation model; and then, the knowledge discovery and system regulation module deduces an energy-saving optimization setting suggestion by using the temperature difference and energy consumption adaptive inference model and the real-time sensing data so as to continuously and adaptively regulate and control the air conditioning system.
Description
Technical field
The present invention relates to a kind of air-conditioning control technique, particularly relate to a kind of air-conditioning control technique carrying out intelligence learning and regulation and control constantly.
Background technology
Improve air-conditioning system and have a lot of method, such as be equipped with the manipulation of diversification compound or replacing, but if control target can be minimized, and import intelligent technology, then can under existing air-conditioning system structure, reach energy conservation object with minimum change scope and minimum manipulation project, save cost with this and avoid the risk of destructive construction.
According to air-conditioning system data results, with regard to internal factor, the power consumption of ice water host computer is the main origin cause of formation, and the frozen water temperature difference is one of great crucial controllable factor, namely the temperature difference more consumes energy more greatly, and can impact the frozen water of air-conditioning system, cooling water, refrigerant three circulations jointly.Therefore, to minimize control target, frozen water pumping of controlling well is just primary work.
Frozen water pumping is as the flow controlling frozen water, the inverter motor frequency setting of frozen water pumping must be higher, frozen water flow will be faster, therefore, the return water temperature of frozen water will more close to the leaving water temperature of its setting, and the frozen water temperature difference will diminish, although such result can allow the power consumption of ice water host computer reduce, but the motor power consumption of frozen water pumping can be caused to increase, and overall power consumption not necessarily can reduce.Contrary, the motor frequency of frozen water pumping sets lower, and rotating speed and frozen water flow all can be slack-off, but, though this measure can save the motor power consumption of frozen water pumping, the power consumption of ice water host computer can be caused to increase.
Therefore, how to try to achieve overall energy consumption minimization, and optimal balance point is obtained between the power consumption of ice water host computer and the motor of frozen water pumping are consumed energy, namely need according to instant situation instantly, suitably use the best setting reaching this usefulness immediately to regulate and control simultaneously.
Summary of the invention
In view of industry now target for reaching, the main object of the present invention is to provide a kind of air-conditioning control technique that can carry out intelligence learning and regulation and control constantly.
In order to achieve the above object and other objects, the invention provides a kind of intelligence learning energy-conserving regulate and control method, comprising: initialization system regulation and control data and instant sense data is extracted to air-conditioning system; According to these set system regulation data and this instant sense data of extracting, carry out empirical learning and logical derivation with adaptive inference modeling techniques, to set up and to encapsulate the temperature difference and power consumption adaptive inference model; And utilize this temperature difference and power consumption adaptive inference model, and from the instant sense data that this air-conditioning system is extracted, reason out energy saving optimizing setting suggestion with data prospecting techniques, and then this air-conditioning system is carried out to the adaptive regulation and control of continuation by knowledge discovery engine technique.
In addition, the present invention also provides a kind of intelligence learning energy-conserving regulate and control system, comprising: system regulation data setting module, regulates and controls data and extract instant sense data to air-conditioning system for initialization system; Empirical learning and logical derivation module, according to set system regulation data and the instant sense data that extracts, carry out empirical learning and logical derivation with adaptive inference modeling techniques, to set up and to encapsulate the temperature difference and power consumption adaptive inference model; And knowledge discovery and system regulation module, for utilizing this temperature difference and power consumption adaptive inference model, and from the instant sense data that this air-conditioning system is extracted, with data prospecting techniques data prospect in module reason out energy saving optimizing setting suggestion, and then by knowledge discovery engine technique to this air-conditioning system carry out continuation adaptive regulation and control.
Compared to prior art, due to intelligence learning energy-conserving regulate and control system and method for the present invention, can routinely utilize the instant temperature difference set up and power consumption adaptive inference model to be optimized set and adaptive regulates and controls, therefore optimal balance point can be obtained between the motor of the power consumption of ice water host computer and frozen water pumping consumes energy, reach overall maximum energy-saving setting.
Accompanying drawing explanation
Fig. 1 is system and the schematic flow sheet of intelligence learning energy-conserving regulate and control system and method for the present invention; And
Fig. 2 is the time diagram that the continuation of carrying out according to the present invention regulates and controls.
Primary clustering symbol description
1 intelligence learning energy-conserving regulate and control system
10 system regulation data setting modules
11 empirical learnings and logical derivation module
12 knowledge discoveries and system regulation module
20 to 23 time sections
A to p program.
Detailed description of the invention
For the benefit of auditor's effect of understanding technical characteristic of the present invention, content and advantage and can reaching, now invention of the present invention is coordinated accompanying drawing, and be described as follows with the expression-form of embodiment, and wherein used accompanying drawing, its purport is only the use of signal and aid illustration, may not be actual proportions after the invention process and precisely configure, therefore should not limit to the interest field of the present invention on reality is implemented with regard to appended accompanying drawing ratio and configuration relation, conjunction be first chatted bright.
Refer to Fig. 1, to understand the intelligence learning energy-conserving regulate and control system and method being applied in air-conditioning system provided by the invention.What need first illustrate is, the present embodiment intelligence learning energy-conserving regulate and control system 1 comprises system regulation data setting module 10, empirical learning and logical derivation module 11, and knowledge discovery and system regulation module 12, to perform intelligence learning energy-conserving regulate and control method provided by the present invention, but, system regulation data setting module 10, empirical learning and logical derivation module 11, and knowledge discovery and system regulation module 12 also optionally can carry out merging or be separated according to different demands and entity and virtual setting, namely, the framework that also can be different from intelligence learning energy-conserving regulate and control system 1 of the present invention implements intelligence learning energy-conserving regulate and control method provided by the invention.First need propose, intelligence learning energy-conserving regulate and control System and method for of the present invention, can be used for carrying out scheduling regulation and control that are instant or non-instant.
Before implementing the present invention, the Schilling air-conditioning system of building in thing can configure the sensor of measurement needed for target data, and make sensor store to rear end servo host or information system platform via communication technology transmission at-once monitor data constantly.The mode of data storing can be database, data warehousing or is archives economy, and can be used as the Data Source of ASSOCIATE STATISTICS of the present invention and analysis.Data content can have a timing, and is the course record of the measuring value at instant scene.And the report control platform of air-conditioning system, instant setting and regulation and control can be carried out to relevant device.
When implementing of the present invention, system regulation data setting module 10 can regulate and control data and extracts instant sense data to air-conditioning system by first initialization system; Then, the instant sense data of the system regulation data that empirical learning and logical derivation module 11 meeting foundation set and extraction, carries out empirical learning and logical derivation with adaptive inference modeling techniques, to set up and to encapsulate the temperature difference and power consumption adaptive inference model; Then, knowledge discovery and system regulation module 12 can utilize this temperature difference and power consumption adaptive inference model, and from the instant sense data that this air-conditioning system is extracted, with data prospecting techniques data prospect in module reason out energy saving optimizing setting suggestion, and then by knowledge discovery engine technique to this air-conditioning system carry out continuation adaptive regulation and control.Described data prospect (Data mining) technology, refer to that automatic searching is hidden in the calculation process wherein having particular associative from a large amount of data, can consult the pertinent literature of the art in detail, no longer repeat further in this.
In one embodiment, the system regulation data that intelligence learning energy-conserving regulate and control system 1 also can regulate and control according to the adaptive of air-conditioning system being carried out to continuation, and the instant sense data of instant sense data is extracted from this air-conditioning system, judge whether to meet modeling conditions again, if, then again activate empirical learning and logical derivation module 11, if not, then again start knowledge discovery and system regulation module 12.Namely this perform the program p of accompanying drawing.
Further it, system regulation data setting module 10, first can set the source of instant sense data, extract target and method, namely perform the program a of accompanying drawing; Reset data time sequence analyst coverage, namely perform the program b of accompanying drawing; The effective data sampling condition of secondary setting, namely performs the program c of accompanying drawing; Then set continuous learning and adaptive adjustment programme, namely perform the program d of accompanying drawing.
Empirical learning and logical derivation module 11, can first extract from the Data Source of setting and filter relevant modeling data, namely perform the program e of accompanying drawing; Be loaded into modeling data again to carry out sized and standardization, namely perform the program f of accompanying drawing; Secondaryly set up temperature difference adaptive inference model with modeling data by adaptive inference modeling techniques, perform the program g of accompanying drawing; More with modeling data by adaptive inference modeling techniques, and according to the inference result of this temperature difference adaptive inference model, set up power consumption adaptive inference model further, perform the program h of accompanying drawing; Secondaryly utilize adaptive inference model comparison technology, the temperature difference relatively previously set up and power consumption adaptive inference model and the most newly-established temperature difference and power consumption adaptive inference model, to select the minimum temperature difference of error and power consumption adaptive inference model, to carry out being loaded into the temperature difference and power consumption adaptive inference model and to encapsulate, namely perform the program i of accompanying drawing.
Knowledge discovery and system regulation module 12, first can be loaded into the temperature difference selected and prospect module (not shown) with power consumption adaptive inference model to data, namely perform the program j of accompanying drawing; Recycling data are prospected module and are extracted instant sense data, namely perform the program k of accompanying drawing; The instant sense data of secondary foundation reasons out multiple power consumption strategy combination with adaptive inference technology, namely performs the program l of accompanying drawing; More advising as setting by selecting the most energy-conservation Optimizing Suggestions of choosing in the plurality of power consumption strategy combination using adaptive inference and optimization technique, namely performing the program m of accompanying drawing; Secondly according to this energy saving optimizing setting suggestion, this air-conditioning system is set, namely perform the program n of accompanying drawing; Then the adaptive of this air-conditioning system row continuation is regulated and controled, namely perform the program o of accompanying drawing.In other words, the present invention, by technology such as adaptive inference, adaptive regulation and control, based on artificial intelligence, and is preserved and transfer specific knowledge, is reached effect of self adaptation (self adapting) further.In addition, described optimization (Optimization) technology is also known as optimisation technique, refer to the calculation process finding a most suitable settling mode under adjoint many restriction and the conflicting environment of condition, the relate art literature of this area can be consulted in detail, no longer repeat further in this.
The embody rule content of aforesaid program a to program p, can consider following exemplary context in light of actual conditions.
In program a, can selected target building and the Back end data servo host belonging to air-conditioning system or data management platform, and provide and can the target data extracted of relevant parameter and the institute wish of Data Source line access therewith set.
In program b, what first can set every sense data takes the unit interval, such as by per hour or per minute in units of, using take the mean value of data within the unit interval as analysis use, and data are when extracting, the unit interval set according to this is hived off; Then can set for carrying out the valid data stroke count scope analyzed, every data are all the result after valid data sampling processing, and in chronological sequence sort, and this range set represents Air-condition system control experience within one section of special time and result, it is also the space that preparation will be carried out systematization empirical learning and adaptive and adjusts.In other words, follow-up process can be set up adaptive inference model to experience interior during this section and is optimized and obtains energy-conservation manipulation knowledge, then regulates and controls the present situation of air-conditioning system based on this.
In program c, can come via expert interviewing and data analysis two kinds of methods, determine air-conditioning system is regulated and controled restriction and the scope being defined to valid data and condition.Such as first from the discussion record of air-conditioning system management and control responsible person or from air-conditioning system description, Dynamic System handbook, the setting restriction that access should be appreciated that when operating air-conditioning system and scope, and the conventional operating habit of corresponding various situation and inside processing specify, especially in the part of instant monitoring objective data.Then obtaining the eligible as setting of valid data via data analysis again, as analyzed via the air-conditioning system historical data scheduling to last at least one year, determining the effective range of specific objective data and the condition of filtering outlier.
In program d, the frequency that continuous learning and adaptive adjust and method can be determined, study scheduling is set according to the temporal frequency of specifying, under set time point, judge whether to relearn the experience in data area that program b defines, and then with the logic that this learns, add data instantly, prospected by data and with knowledge discovery technology, adaptive is done again to air-conditioning system and regulate and control.In other words, down-stream can be come regularly to perform correlation step according to scheduling set herein.And whether to enter the learning program re-establishing new adaptive inference model immediately, then need to set herein, and as the Rule of judgment of program p.
In program e, the line of originating with target data can be set up via the setting done in program a, and process every target data.The extraction of values of every item number certificate is for according to the mean value in the unit interval set by program b.Surely the condition ordered according to program c is again carried out screening to target data and is filtered, to obtain the valid data record stroke count scope that the b that is in order defines, to carry out follow-up correlation analysis process.
In program f, adaptive inference model modelling approach and the model data process of adaptive inference and loading can be comprised.Such as, inference model can be formed by the construction of algorithm institute, via input and export data and carry out adaptive study, for training the logical relation between these two groups of data, is used for description experience being remembered; Algorithm model, after foundation completes, can carry out inference according to existing experience to input data, and draw rational Output rusults; The algorithm composition of adaptive inference model can use various ways to realize, and comprises algorithm with regress analysis method, class neural algorithm, fuzzy logic algorithm and decision Tree algorithms etc.In addition, valid data, before loading adaptive inference model, according to the characteristic of data or model algorithm, can do sized and standardized process to data, and then are supplied to model algorithm and carry out training study; The sized codomain scope of initial data that provides of data is changed, make it be applicable to being standardized and can accept by adaptive inference model algorithm; Different data samples can be carried out normality distributionization by data normalization, can carry out analyzing and processing under same benchmark; Same, before performing and carrying out logical deduction by the model after completing training study, also need this road handling procedure; And data after completion processing, be loaded in empirical learning and logical deduction module 11 by it.
In program g, temperature difference adaptive inference model can be first model be established in empirical learning and logical deduction module 11, for adaptive learn in designated duration air-conditioning setting and operating experience, find out frozen water end pressure reduction, frozen water pumping setpoint frequency, cooling water pumping setpoint frequency and the cooling water Water Exit temperature difference, to the interact relation that the frozen water Water Exit temperature difference causes, and set up out effective inferrible model for its logic.
In program h, power consumption adaptive inference model can be the model that and then empirical learning and logical deduction module 11 should be established out after setting up temperature difference adaptive inference model, for adaptive learn in designated duration air-conditioning setting and operating experience, find out frozen water end pressure reduction, frozen water pumping setpoint frequency, cooling water pumping setpoint frequency, the cooling water Water Exit temperature difference and the frozen water Water Exit temperature difference that obtains through inference sums up to target power consumption the interact relation caused, and set up out effective inferrible model for its logic.
In program i, data can carry out correlation analysis in scope set by service routine b, the adaptive inference model newly built up with comparison program g and program h, with the error mean square root (RMS of the model previously established, Root ofMean Square), and retain the less model of error mean square root, and give up error mean square root the greater.
In program j, can prospecting in module by being loaded into data by two the adaptive inference models selected, preparing that data are carried out to data with this and prospecting, the potential knowledge outside the best setting included with the experience of excavating or experience.
In program k, can extract the instant sense data of time point instantly, such as frozen water end pressure reduction, cooling water pumping setpoint frequency, cooling water leaving water temperature or cooling water enter coolant-temperature gage.
In program l, the instant sense data of current time point can be used, and the CHP frequency range setting that loader c does, then possible input combination is generated.Then with temperature difference adaptive inference model, according to the experience logic that learn of institute, reasoning out the frozen water temperature difference of each combination, and then transfer to power consumption adaptive inference model, summing up according to going the target reasoning out each combination to consume energy in the past experience in institute's range of definition.The numerical result obtained from logical deduction model, can again through anti-sized and anti-standardization, with restoring data.Described sized (scaling) refers to the calculation process of scale, anti-sized (Rescaling) refers to the calculation process mapping back true yardstick, standardization (Standardization) refers to formulation standard and the calculation process of reaching an agreement with regard to it, the pertinent literature of the art can be consulted in detail, no longer repeat further in this.After inference work completes, namely the strategy matrix of constrained input is completed.
In program m, from the strategy matrix that program l completes, in various possible constrained input strategy, the CHP setpoint frequency that target power consumption sum total can be facilitated minimum can be found.In program n, can by drawn new CHP setpoint frequency, do air-conditioning system and control setting adjustment, make it run meeting in energy saving condition instantly.In program o, scheduling can be carried out and judge and management and control, if time point instantly meets frequency set in program d, then enter next program.Otherwise, then continue standby.
In program p, can judge whether to enter the learning program re-establishing new adaptive inference model immediately from the setting that program d does.If so, namely restart empirical learning and logical derivation module 11 and knowledge discovery and system regulation module 12, namely performing a programme e is to program o.If not, then representative can continue to adopt old adaptive inference model thus braking knowledge discovery and system regulation module 12, namely performs previous program j to program o.Certainly, program p allows the sub-determining program increasing other newly, if be judged to be eligible through new sub-determining program, then and the adaptive inference model that can enter again modeling program or use specific outside to import.
Namely Fig. 2 illustrates the schematic diagram performing the continuation adaptive regulation and control that continuation of the present invention is reached.In the accompanying drawings, time section 20,21, represents and performs intelligence learning energy-conserving regulate and control method of the present invention for the first time, namely represents and starts intelligence learning energy-conserving regulate and control system 1 of the present invention for the first time; Time section 22,23, represents second time and performs intelligence learning energy-conserving regulate and control method of the present invention, namely represents second time and starts intelligence learning energy-conserving regulate and control system 1 of the present invention, by that analogy.In addition, the continuation that time section 20,22 can represent execution technology of the present invention to be provided is detected, learns and is derived, and time section 21,23 then can represent the instant regulation and control performing technology of the present invention and carry out air-conditioning system.It can thus be appreciated that, n-th time to air-conditioning system complete instant adaptive regulation and control instantly, intelligence learning energy-conserving regulate and control system and method for the present invention can be in carry out (n+1)th time continuation detecting, study with derivation process in.
In addition, aforementioned this air-conditioning system to be regulated and controled, can refer to regulate and control the ice water host computer side of this air-conditioning system, such as, regulate and control frozen water pump frequency.Initialization system regulation and control data, can comprise setting record time-histories, frozen water pump frequency and cooling water pump frequency.Extract instant sense data, can comprise and extract that frozen water enters coolant-temperature gage, frozen water leaving water temperature, cooling water enter coolant-temperature gage, cooling water leaving water temperature, frozen water flow, frozen water end pressure reduction, frozen water pumping consumed power, cooling water pumping consumed power and ice water host computer consumed power.And energy saving optimizing setting suggestion, can comprise and carry out suggestion corresponding to frozen water pumping consumed power, cooling water pumping consumed power and ice water host computer consumed power.
Compared to prior art, due to intelligence learning energy-conserving regulate and control system and method for the present invention, routinely can utilize the instant temperature difference set up and power consumption adaptive inference model, carry out energy saving optimizing setting for air-conditioning system to regulate and control with instant adaptive, therefore instant optimal balance point can be tried to achieve between the motor of the power consumption of ice water host computer and frozen water pumping consumes energy, and then reach the maximum energy-saving setting of air-conditioning system entirety, and avoid the disadvantages of prior art.
After carrying out actual experiment, compared to the front and back same January of 2 years, use technology of the present invention, can effectively promote the amount of energy saving of air-conditioning, refrigerating capacity (RT), with efficiency ratio (EER), its promote such as illustrated in following relevant comparison sheet:
Table 1: the amount of energy saving comparison sheet in 2 years identical months of front and back
Table 2: the refrigerating capacity enhancing rate comparison sheet in 2 years identical months of front and back
Table 3: the Energy Efficiency Ratio enhancing rate comparison sheet in 2 years identical months of front and back
Above-described embodiment only for illustrative principle of the present invention and effect thereof, but not for limiting the present invention.Those skilled in the art all without prejudice under spirit of the present invention and category, can modify to above-described embodiment.Therefore the rights protection scope of this case, should listed by claims.
Claims (20)
1. an intelligence learning energy-conserving regulate and control method, is characterized in that, comprising:
1) initialization system regulates and controls data and extracts instant sense data to air-conditioning system;
2) according to these set system regulation data and this instant sense data of extracting, empirical learning and logical derivation is carried out with adaptive inference modeling techniques, to set up and to encapsulate the temperature difference and power consumption adaptive inference model; And
3) this temperature difference and power consumption adaptive inference model is utilized, and from the instant sense data that this air-conditioning system is extracted, reason out energy saving optimizing setting suggestion with data prospecting techniques, and then this air-conditioning system is carried out to the adaptive regulation and control of continuation by knowledge discovery engine technique.
2. intelligence learning energy-conserving regulate and control method according to claim 1, it is characterized in that, the method also comprises step 4), according to the system regulation data that the adaptive of this air-conditioning system being carried out to continuation regulates and controls, and from the instant sense data that this air-conditioning system is extracted, judge whether to meet modeling conditions again, if, then proceed to step 2), if not, then proceed to step 3).
3. intelligence learning energy-conserving regulate and control method according to claim 1, it is characterized in that, this step 1) also comprise: set the source of this instant sense data, extract target and method, setting data Time-Series analysis scope, set effective data sampling condition, and set continuous learning and adaptive adjustment programme.
4. intelligence learning energy-conserving regulate and control method according to claim 3, is characterized in that, this step 2) further comprising the steps of:
2.1) extract from the source of the instant sense data of this setting and filter relevant modeling data;
2.2) this modeling data is loaded into carry out sized and standardization;
2.3) temperature difference adaptive inference model is set up with this modeling data by adaptive inference modeling techniques;
2.4) with this modeling data by this adaptive inference modeling techniques, and according to the inference result of this temperature difference adaptive inference model, set up power consumption adaptive inference model further; And
2.5) this adaptive inference model comparison technology is utilized, the temperature difference relatively previously set up and power consumption adaptive inference model and the most newly-established temperature difference and power consumption adaptive inference model, to select the minimum temperature difference of error and power consumption adaptive inference model, to carry out being loaded into and to encapsulate.
5. intelligence learning energy-conserving regulate and control method according to claim 4, is characterized in that, step 2.3) further comprising the steps of:
2.3.1) temperature difference selected and power consumption adaptive inference model is loaded into;
2.3.2) this instant sense data is extracted;
2.3.3) multiple power consumption strategy combination is reasoned out according to this instant sense data with adaptive inference technology;
2.3.4) suggestion is set by selecting the most energy-conservation Optimizing Suggestions of choosing in the plurality of power consumption strategy combination as this energy saving optimizing using adaptive inference and optimization technique;
2.3.5) according to this energy saving optimizing setting suggestion, this air-conditioning system is set; And
2.3.6) adaptive of this air-conditioning system row continuation is regulated and controled.
6. intelligence learning energy-conserving regulate and control method according to claim 1, is characterized in that, this air-conditioning system is carried out to the adaptive regulation and control of continuation, refers to and regulate and control the ice water host computer side of this air-conditioning system.
7. intelligence learning energy-conserving regulate and control method according to claim 6, is characterized in that, regulates and controls the ice water host computer side of this air-conditioning system, refers to regulation and control frozen water pump frequency.
8. intelligence learning energy-conserving regulate and control method according to claim 1, is characterized in that, this initialization system regulation and control data comprise setting record time-histories, frozen water pump frequency and cooling water pump frequency.
9. intelligence learning energy-conserving regulate and control method according to claim 1, it is characterized in that, the instant sense data of this extraction comprises and extracts that frozen water enters coolant-temperature gage, frozen water leaving water temperature, cooling water enter coolant-temperature gage, cooling water leaving water temperature, frozen water flow, frozen water end pressure reduction, frozen water pumping consumed power, cooling water pumping consumed power and ice water host computer consumed power.
10. intelligence learning energy-conserving regulate and control method according to claim 1, is characterized in that, this energy saving optimizing setting suggestion comprises corresponding to frozen water pumping consumed power, cooling water pumping consumed power and ice water host computer consumed power.
11. 1 kinds of intelligence learning energy-conserving regulate and control systems, is characterized in that, comprising:
System regulation data setting module, regulates and controls data for initialization system and extracts instant sense data to air-conditioning system;
Empirical learning and logical derivation module, according to set system regulation data and the instant sense data that extracts, carry out empirical learning and logical derivation with adaptive inference modeling techniques, to set up and to encapsulate the temperature difference and power consumption adaptive inference model; And
Knowledge discovery and system regulation module, for utilizing this temperature difference and power consumption adaptive inference model, and from the instant sense data that this air-conditioning system is extracted, with data prospecting techniques data prospect in module reason out energy saving optimizing setting suggestion, and then by knowledge discovery engine technique to this air-conditioning system carry out continuation adaptive regulation and control.
12. intelligence learning energy-conserving regulate and control systems according to claim 11, it is characterized in that, this intelligence learning energy-conserving regulate and control system, also according to the system regulation data that the adaptive of this air-conditioning system being carried out to continuation regulates and controls, and from the instant sense data that this air-conditioning system is extracted, judge whether to meet modeling conditions again, if, then again activate this empirical learning and logical derivation module, if not, then again start this knowledge discovery and system regulation module.
13. intelligence learning energy-conserving regulate and control systems according to claim 11, is characterized in that, this system regulation data setting module, first set the source of this instant sense data, extract target and method; Reset data time sequence analyst coverage; The effective data sampling condition of secondary setting; Then continuous learning and adaptive adjustment programme is set.
14. intelligence learning energy-conserving regulate and control systems according to claim 13, is characterized in that, this empirical learning and logical derivation module, first extract from the source of the instant sense data of this setting and filter relevant modeling data; Be loaded into this modeling data again to carry out sized and standardization; Secondaryly set up temperature difference adaptive inference model with this modeling data by adaptive inference modeling techniques; More with this modeling data by this adaptive inference modeling techniques, and according to the inference result of this temperature difference adaptive inference model, set up power consumption adaptive inference model further; Next utilizes adaptive inference model comparison technology, and the temperature difference relatively previously set up and power consumption adaptive inference model and the most newly-established temperature difference and power consumption adaptive inference model, to select the minimum temperature difference of error with power consumption adaptive inference model to carry out being loaded into and to encapsulate.
15. intelligence learning energy-conserving regulate and control systems according to claim 14, is characterized in that, this knowledge discovery and system regulation module, are first loaded into the temperature difference selected and prospect module with power consumption adaptive inference model to these data; Recycle these data to prospect module and extract this instant sense data; Secondly multiple power consumption strategy combination is reasoned out according to this instant sense data with adaptive inference technology; More set suggestion by selecting the most energy-conservation Optimizing Suggestions of choosing in the plurality of power consumption strategy combination as energy saving optimizing using adaptive inference and optimization technique; Again according to this energy saving optimizing setting suggestion, this air-conditioning system is set; Then the adaptive of this air-conditioning system row continuation is regulated and controled.
16. intelligence learning energy-conserving regulate and control systems according to claim 11, is characterized in that, this air-conditioning system are carried out to the adaptive regulation and control of continuation, refer to and regulate and control the ice water host computer side of this air-conditioning system.
17. intelligence learning energy-conserving regulate and control systems according to claim 16, is characterized in that, regulate and control the ice water host computer side of this air-conditioning system, refer to regulation and control frozen water pump frequency.
18. intelligence learning energy-conserving regulate and control systems according to claim 11, is characterized in that, this initialization system regulation and control data comprise setting record time-histories, frozen water pump frequency and cooling water pump frequency.
19. intelligence learning energy-conserving regulate and control systems according to claim 11, it is characterized in that, the instant sense data of this extraction comprises and extracts that frozen water enters coolant-temperature gage, frozen water leaving water temperature, cooling water enter coolant-temperature gage, cooling water leaving water temperature, frozen water flow, frozen water end pressure reduction, frozen water pumping consumed power, cooling water pumping consumed power and ice water host computer consumed power.
20. intelligence learning energy-conserving regulate and control systems according to claim 11, is characterized in that, this energy saving optimizing setting suggestion corresponds to frozen water pumping consumed power, cooling water pumping consumed power and ice water host computer consumed power.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW102145305 | 2013-12-10 | ||
| TW102145305A TWI516886B (en) | 2013-12-10 | 2013-12-10 | Intelligent learning energy-saving control system and method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104697107A true CN104697107A (en) | 2015-06-10 |
| CN104697107B CN104697107B (en) | 2017-03-29 |
Family
ID=53344492
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310724423.3A Active CN104697107B (en) | 2013-12-10 | 2013-12-24 | Intelligent learning energy-saving regulation and control system and method |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN104697107B (en) |
| TW (1) | TWI516886B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109357352A (en) * | 2018-11-05 | 2019-02-19 | 四川长虹电器股份有限公司 | Air conditioner intelligent temperature control system and method based on Decision Tree Algorithm |
| WO2020056960A1 (en) * | 2018-09-19 | 2020-03-26 | 珠海格力电器股份有限公司 | Temperature control method and apparatus for unit, and air conditioning unit |
| CN112032982A (en) * | 2020-08-21 | 2020-12-04 | 深圳市建滔科技有限公司 | Indoor environment comfort level improving method based on co-policy Monte Carlo algorithm |
| TWI751502B (en) * | 2020-02-25 | 2022-01-01 | 高齊能源科技股份有限公司 | Performance evaluation method of ice water main engine |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI644062B (en) | 2017-06-26 | 2018-12-11 | 群光電能科技股份有限公司 | Adjusting system |
| CN114303161A (en) * | 2019-09-06 | 2022-04-08 | 三菱电机株式会社 | Learning device, learning method, learning data generation device, learning data generation method, inference device, and inference method |
| TWI778329B (en) | 2020-02-21 | 2022-09-21 | 香港商希瑞科技股份有限公司 | Methods and systems for information storage |
| TWI771882B (en) | 2021-01-27 | 2022-07-21 | 群光電能科技股份有限公司 | Power converter device and driving method |
| TWI790917B (en) * | 2022-02-11 | 2023-01-21 | 中國鋼鐵股份有限公司 | Method for estimating power consumption of water chiller and cooling system |
| TWI795279B (en) * | 2022-04-27 | 2023-03-01 | 中國鋼鐵股份有限公司 | A method for regulating and controlling energy-saving operation of a chiller machine |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102003772A (en) * | 2010-11-30 | 2011-04-06 | 中国建筑西南设计研究院有限公司 | Energy-saving optimized control method of water source heat pump |
| CN102012077A (en) * | 2010-12-06 | 2011-04-13 | 北京星达技术开发公司 | Energy-saving control system and control method of central air conditioning freezing station |
| KR20110100895A (en) * | 2010-03-05 | 2011-09-15 | 삼성물산 주식회사 | The building mutual assistance control method which uses an optimization energy management system |
| EP2634498A1 (en) * | 2010-10-27 | 2013-09-04 | Technomirai Co., Ltd | Air conditioning control system and program |
| CN103322645A (en) * | 2013-06-13 | 2013-09-25 | 浙江工业大学 | Predictive control method for return water temperature of chilled water of central air-conditioner |
-
2013
- 2013-12-10 TW TW102145305A patent/TWI516886B/en active
- 2013-12-24 CN CN201310724423.3A patent/CN104697107B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20110100895A (en) * | 2010-03-05 | 2011-09-15 | 삼성물산 주식회사 | The building mutual assistance control method which uses an optimization energy management system |
| EP2634498A1 (en) * | 2010-10-27 | 2013-09-04 | Technomirai Co., Ltd | Air conditioning control system and program |
| CN102003772A (en) * | 2010-11-30 | 2011-04-06 | 中国建筑西南设计研究院有限公司 | Energy-saving optimized control method of water source heat pump |
| CN102012077A (en) * | 2010-12-06 | 2011-04-13 | 北京星达技术开发公司 | Energy-saving control system and control method of central air conditioning freezing station |
| CN103322645A (en) * | 2013-06-13 | 2013-09-25 | 浙江工业大学 | Predictive control method for return water temperature of chilled water of central air-conditioner |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020056960A1 (en) * | 2018-09-19 | 2020-03-26 | 珠海格力电器股份有限公司 | Temperature control method and apparatus for unit, and air conditioning unit |
| CN109357352A (en) * | 2018-11-05 | 2019-02-19 | 四川长虹电器股份有限公司 | Air conditioner intelligent temperature control system and method based on Decision Tree Algorithm |
| CN109357352B (en) * | 2018-11-05 | 2021-01-26 | 四川长虹电器股份有限公司 | Air conditioner intelligent temperature control system and method based on decision tree classification algorithm |
| TWI751502B (en) * | 2020-02-25 | 2022-01-01 | 高齊能源科技股份有限公司 | Performance evaluation method of ice water main engine |
| CN112032982A (en) * | 2020-08-21 | 2020-12-04 | 深圳市建滔科技有限公司 | Indoor environment comfort level improving method based on co-policy Monte Carlo algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104697107B (en) | 2017-03-29 |
| TWI516886B (en) | 2016-01-11 |
| TW201523174A (en) | 2015-06-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104697107A (en) | Intelligent learning energy-saving regulation and control system and method | |
| Xue et al. | Fault detection and operation optimization in district heating substations based on data mining techniques | |
| CN103853106B (en) | A kind of energy consumption Prediction Parameters optimization method of building energy supplied equipment | |
| Xiao et al. | Data mining in building automation system for improving building operational performance | |
| Sun et al. | Gradual fault early stage diagnosis for air source heat pump system using deep learning techniques | |
| CN108510006A (en) | A kind of analysis of business electrical amount and prediction technique based on data mining | |
| CN102789220B (en) | System energy consumption process management device and system energy consumption process management control method | |
| CN103912966A (en) | Optimal control method for ground source heat pump refrigerating system | |
| CN101021914A (en) | Heating ventilating and air conditioner load predicting method and system | |
| Sun et al. | Optimization of support vector regression model based on outlier detection methods for predicting electricity consumption of a public building WSHP system | |
| CN117421618B (en) | Building energy consumption monitoring method and system | |
| CN112884012A (en) | Building energy consumption prediction method based on support vector machine principle | |
| CN112700085A (en) | Association rule based method, system and medium for optimizing steady-state operation parameters of complex system | |
| CN112288157A (en) | Wind power plant power prediction method based on fuzzy clustering and deep reinforcement learning | |
| Gao et al. | How to improve the application potential of deep learning model in HVAC fault diagnosis: Based on pruning and interpretable deep learning method | |
| Tian et al. | An unsupervised data mining-based framework for evaluation and optimization of operation strategy of HVAC system | |
| Liu et al. | Tomato detection based on convolutional neural network for robotic application | |
| CN104091042A (en) | Operating data-based method for energy conservation diagnosis and energy conservation potential analysis on central air-conditioning system | |
| CN106447207A (en) | Green building pre-evaluation system based on large data analysis | |
| CN113434989A (en) | Pipe network leakage amount prediction method and system based on attention mechanism and LSTM | |
| CN117113086A (en) | Energy storage unit load prediction method, system, electronic equipment and medium | |
| CN110762768A (en) | Energy efficiency ratio prediction method and device for refrigeration host of central air-conditioning system | |
| Moreno et al. | Exploiting IoT-based sensed data in smart buildings to model its energy consumption | |
| Cheng et al. | Abnormal energy consumption detection using ensemble model for water chilling unit on HVAC system | |
| CN113959071A (en) | Machine learning assistance-based operation control optimization method for centralized water chilling unit air conditioning system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |