CN113902582A - Building comprehensive energy load prediction method and system - Google Patents
Building comprehensive energy load prediction method and system Download PDFInfo
- Publication number
- CN113902582A CN113902582A CN202110990133.8A CN202110990133A CN113902582A CN 113902582 A CN113902582 A CN 113902582A CN 202110990133 A CN202110990133 A CN 202110990133A CN 113902582 A CN113902582 A CN 113902582A
- Authority
- CN
- China
- Prior art keywords
- time
- load
- indoor
- energy
- outdoor
- 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
- 238000000034 method Methods 0.000 title claims abstract description 35
- 230000007613 environmental effect Effects 0.000 claims abstract description 37
- 238000012706 support-vector machine Methods 0.000 claims abstract description 29
- 238000012937 correction Methods 0.000 claims abstract description 9
- 238000005265 energy consumption Methods 0.000 claims description 35
- 239000013598 vector Substances 0.000 claims description 27
- 238000001816 cooling Methods 0.000 claims description 13
- 238000012544 monitoring process Methods 0.000 claims description 10
- 230000005855 radiation Effects 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 9
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 6
- 238000010438 heat treatment Methods 0.000 claims description 5
- 230000015654 memory Effects 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000005096 rolling process Methods 0.000 claims description 3
- 230000006399 behavior Effects 0.000 description 5
- 230000005611 electricity Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 230000007704 transition Effects 0.000 description 4
- 239000002699 waste material Substances 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 230000004888 barrier function Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000003064 k means clustering Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000005192 partition Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000004138 cluster model Methods 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000013486 operation strategy Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000005057 refrigeration Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Tourism & Hospitality (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a building comprehensive energy load prediction method and a system, comprising the following steps: dividing a building into different functional areas; establishing an indoor environment standard mode set aiming at different functional areas; clustering environmental data and personnel information in each functional area, and performing benchmarking correction on the environmental data and the personnel information with a standard mode set to form an energy use habit set of each functional area; constructing a support vector machine prediction model based on outdoor meteorological environment data, energy utilization habits and personnel arrangement information parameters; and obtaining a load prediction result of a prediction day based on the support vector machine prediction model. The accuracy of building multi-load prediction is effectively improved, the building energy-saving operation can be guided, and the building energy management level is improved.
Description
Technical Field
The invention belongs to the technical field of comprehensive energy load prediction, and particularly relates to a building comprehensive energy load prediction method and system.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
The comprehensive energy system is used as a novel energy system integrating 'source-network-load-storage', has the advantages of energy conservation, environmental protection, economy and the like, and is more and more widely applied to building energy supply scenes. The load prediction can be used for building energy-saving operation or directly provides load basis and data basis for the design and the optimized operation of a comprehensive energy system, the accurate prediction of the load is an important basis for realizing the optimized operation of the energy system, and the accurate prediction of the cold, heat and electricity energy utilization load of the building has great significance for energy consumption management and control, energy saving and cost reduction.
At present, a lot of research has been carried out by scholars at home and abroad on the load prediction technology, including a heat transfer simulation calculation method based on a building structure, a prediction method based on mathematical statistics, time series prediction, a linear regression prediction method and the like. As the cold and heat load of the building is influenced by factors such as weather, holidays, social activities, the number of users and habits, the method has the characteristics of strong nonlinearity and randomness, and the problems of complex realization, low prediction precision and the like generally exist in the conventional prediction technology. With the increasing maturity of computer technology, methods of data mining and machine learning are endless, and a new idea is provided for the research of load prediction technology. Perceptron, artificial neural network, XGboost model, intelligent algorithm and the like are gradually applied to the field of load prediction to improve the accuracy of load prediction.
The inventor finds that researchers provide the method for realizing the day-ahead prediction of the multi-element load of the comprehensive energy system by considering the internal coupling characteristic of the multi-element load and utilizing a long-term and short-term memory neural network model; researchers also propose to utilize a convolutional neural network to mine internal characteristics of load data and then realize the day-ahead prediction of the load through a deep learning model.
Although the research focuses on the internal characteristics of the data, the research mainly aims at the load data, and the actual indoor environmental conditions and energy consumption behaviors of the energy consumption side are not considered comprehensively, so that on one hand, more accurate load prediction cannot be realized, and on the other hand, energy waste caused by unreasonable energy consumption behaviors cannot be reflected.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a building comprehensive energy load prediction method, which effectively improves the precision of building multi-element load prediction and guides building energy-saving operation by mining the user energy habits through the indoor environment and personnel data.
In order to achieve the above object, one or more embodiments of the present invention provide the following technical solutions:
in a first aspect, a building comprehensive energy load prediction method is disclosed, which comprises the following steps:
dividing a building into different functional areas;
establishing an indoor environment standard mode set aiming at different functional areas;
clustering environmental data and personnel information in each functional area, and performing benchmarking correction on the environmental data and the personnel information with a standard mode set to form an energy use habit set of each functional area;
constructing a support vector machine prediction model based on outdoor meteorological environment data, energy utilization habits and personnel arrangement information parameters;
and obtaining a load prediction result of a prediction day based on the support vector machine prediction model.
The further technical scheme also comprises the following steps: and monitoring indoor environment parameters and load information in real time, comparing the indoor environment standard mode set with the load prediction result, and sending out energy consumption abnormity early warning when the environmental parameter adjustment is over-limit and the load deviation value is greater than a preset threshold value.
According to the further technical scheme, when the building is divided into different functional areas, the functional areas are divided according to the application attributes and the energy utilization characteristics of the different areas of the building.
In a further technical scheme, the indoor environment standard mode set comprises possible environment modes of the functional area in 24 hours per day, and indoor temperature, humidity, illuminance and CO corresponding to each mode2The expected concentration value, the upper and lower adjustment limits thereof and the corresponding running time of each mode.
The further technical scheme comprises the step of obtaining indoor environment data and personnel information, outdoor meteorological environment data and cold and heat power utilization load data of different functional areas.
According to a further technical scheme, the indoor environment data and the personnel information comprise indoor time-by-time air temperature, indoor time-by-time relative humidity, indoor time-by-time illuminance and indoor time-by-time CO2Concentration, indoor time-by-time personnel number and corresponding environment modes;
the outdoor meteorological environment data comprise outdoor weather types, outdoor current day highest temperature, outdoor current day lowest temperature, outdoor time-by-time relative humidity and outdoor time-by-time solar total radiation;
the energy consumption load data includes time-by-time cooling/heating load and time-by-time power consumption of the functional area.
According to the further technical scheme, indoor and outdoor environmental parameters and load hourly data are hourly average values, and are obtained by calculation through an accumulative average method on the basis of second-level historical data; the cold/heat load at a certain moment is obtained by calculating the water supply flow of the air conditioner and the temperature difference of the supplied and returned water.
According to the further technical scheme, the mode of forming the habit set for each functional area is as follows:
clustering indoor environment parameter historical data in the same energy supply cycle and similar time periods in the same period of the previous year in the same mode time by time aiming at the given time period of each functional area to obtain a user energy consumption habit feature vector;
and then, comparing the obtained user energy use habit feature vector with an environment standard pattern set in the functional area, performing limitation correction on parameter components exceeding the upper and lower adjustment limits, and combining to form the user energy use habit set of the functional area.
In a further technical scheme, the step of constructing the support vector machine prediction model is as follows:
selecting N moments with the minimum weighted Euclidean distance as moments to be predicted to construct a training sample set according to the energy consumption characteristic vector of the predicted moment;
normalizing the training samples to eliminate the influence of different dimensions;
optimizing the model parameters of the support vector machine to obtain a prediction model of the support vector machine;
in a further technical scheme, the construction of the energy consumption characteristic vector at the prediction moment specifically comprises the following steps:
constructing a t-period energy consumption characteristic vector X (t) ([ x1, x2, … … and x 12) for the day to be predicted]X 1-x 6 respectively correspond to the outdoor current day weather type value, the outdoor current day maximum temperature, the outdoor current day minimum temperature, the outdoor t-time relative humidity and the outdoor t-time solar total radiation, and x 7-x 10 respectively correspond to the indoor t-time temperature, the indoor t-time relative humidity, the indoor t-time illuminance and the indoor t-time CO corresponding to the habit-enabling feature vector for the user in the functional area2Concentrations, x11, x12 correspond to the expected number of people and environmental patterns at t in the room.
According to the further technical scheme, during prediction, the cold and heat and the electric load are predicted in a time-by-time rolling mode aiming at each functional area, and an energy utilization load curve of a prediction day is obtained.
In a second aspect, a building integrated energy load prediction system is disclosed, comprising:
an indoor environment standard pattern set establishment module configured to: dividing a building into different functional areas, and establishing an indoor environment standard mode set aiming at each different functional area;
each functional area energy use habit set establishing module is configured to: clustering environmental data and personnel information in each functional area, and performing benchmarking correction on the environmental data and the personnel information with a standard mode set to form an energy use habit set of each functional area;
a support vector machine prediction model construction module configured to: constructing a support vector machine prediction model based on outdoor meteorological environment data, energy utilization habits and personnel arrangement information parameters;
and obtaining a load prediction result of a prediction day based on the support vector machine prediction model.
The further technical scheme also comprises the following steps: a real-time monitoring module configured to: and monitoring indoor environment parameters and load information in real time, contrasting an indoor environment standard pattern set and a load prediction result, and sending out energy consumption abnormity early warning when the environmental parameter adjustment is over-limit and the load deviation value is greater than a preset threshold value.
The above one or more technical solutions have the following beneficial effects:
the invention is suitable for multi-energy load prediction and energy-saving operation of buildings or building clusters. The method for predicting the building comprehensive energy load and auditing on line fully considers the influence of weather, building application, the number of users and user energy habits on the multi-element energy load, carries out user energy habit portrayal aiming at different functional areas in the building through a clustering method, corrects the unreasonable energy consumption behaviors, and then constructs a load prediction model by machine learning, thereby effectively improving the precision of the building multi-element load prediction, guiding the building energy-saving operation and improving the building energy management level.
Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
FIG. 1 is a flow chart of a method according to an embodiment of the present invention.
Detailed Description
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention.
The embodiments and features of the embodiments of the present invention may be combined with each other without conflict.
With the updating of the acquisition equipment, especially the application of the internet of things terminal, the acquisition of indoor environment parameters, personnel conditions and the like of the building becomes easier, and conditions are provided for more accurate load prediction and online energy consumption analysis.
Example one
Referring to fig. 1, the embodiment discloses a method for predicting comprehensive energy load of a building, which includes the following steps:
step one, dividing functional areas according to application attributes and energy utilization characteristics of different areas of a building;
first, each area inside the building is divided into 8 types of functional areas, such as office, conference, dining, shopping, apartment, entertainment, data center, and others, according to different application layouts and energy-using characteristics, and each type may include a plurality of functional areas, for example, a conference type may include a plurality of conference functional areas. Considering the convenience of obtaining the metering historical data of the energy utilization system, the division of the functional areas is generally unified with the metering subareas of the energy utilization system.
The functional areas are areas with different application attributes or characteristics, and the considered factors are mainly as follows: the indoor environment standard and the energy consumption behavior habit are different, and the other is that the load change trend is different, namely the rules contained in the historical data are different, so that the prediction precision can be improved through the partition processing.
Step two, establishing an indoor environment standard mode set aiming at different functional areas;
the set of environmental standard patterns in the functional compartment comprises possible environmental patterns of the functional compartment in 24 hours per day, and indoor temperature, humidity, illuminance and CO corresponding to each pattern2The expected concentration value, the upper and lower adjustment limits thereof, the corresponding running time of each mode and the like.
The method comprises the following specific implementation steps: the indoor environment standard mode is distinguished in cooling season, heating season and transition season, and is determined according to the requirements of different functional areas, for example, the standard mode of the office functional area comprises a pre-cooling mode before the cooling season, a cooling mode between the cooling seasons, a cooling mode for people who leave the office during the cooling season, and a cooling stopping mode during the cooling season;
a preheating mode before the heat supply season, a heat supply mode between the heat supply seasons, a heat supply mode for few people during the heat supply season, and a stop mode during the heat supply season;
a transition season on duty mode, a transition season off duty less mode, a transition season off duty unmanned mode and the like.
The data center functional area has a single mode, including a working mode and a non-working mode.
And the other types of functional areas refer to the mode division standard, and the value of the environmental parameter corresponding to each mode is set according to the relevant building environment design standard and the specific operation requirement.
Acquiring indoor environment data and personnel information, outdoor meteorological environment data and cold and heat power utilization load data of different functional areas; the data obtained here is used for cluster analysis and model training in the subsequent steps;
the indoor environment data and personnel information include indoor time-by-time air temperature, indoor time-by-time relative humidity, indoor time-by-time illuminance, and indoor time-by-time CO2Concentration, indoor time-by-time personnel number and corresponding environment modes; the corresponding environment mode is generally a data item in historical data and a history record set manually;
the outdoor meteorological environment data comprise outdoor weather types, outdoor current day highest temperature, outdoor current day lowest temperature, outdoor time-by-time relative humidity and outdoor time-by-time solar total radiation;
the energy utilization load data comprises time-by-time cold/heat loads (cold load in cold season and heat load in hot season) and time-by-time electricity consumption of the functional area, the time-by-time electricity consumption does not contain electricity consumption of air conditioner energy supply equipment, and the electricity consumption of the air conditioner equipment can be directly calculated according to an air conditioner equipment regulation and control operation strategy without prediction.
The indoor and outdoor environmental parameters and the time-by-time load data are time-by-time average values, and the time-by-time trend can be accurately reflected by calculating by an accumulative average method on the basis of second-level historical data; the cold/heat load at a certain moment can be obtained by calculating the water supply flow of the air conditioner and the temperature difference of the supplied and returned water.
The obtained data needs to be cleaned, error values are removed, and missing data are filled.
Fourthly, based on a K-means clustering method, performing benchmarking on the user habit images of each functional area and the standard pattern set to form an energy habit set of each functional area;
and aiming at the given time interval of each functional area, clustering the historical data of the indoor environment parameters in the same mode in the same energy supply cycle and the similar time interval in the same period in the previous year by using K-means clustering time by time, determining the optimal K value through an elbow rule, and obtaining the mass center vectors of K clusters as the user habit-using characteristic vectors. And then, comparing the obtained user energy use habit feature vector with an environment standard pattern set in the functional area, performing limitation correction on parameter components exceeding the upper and lower adjustment limits, and combining to form the user energy use habit set of the functional area.
Specifically, the user habit-enabled feature vector represents the indoor parameters corresponding to the user habit, and may be represented as [ p1, p2, p3, p4, p5 ]]Representative of ambient mode, indoor temperature, humidity, illuminance and CO2And (4) concentration.
Comparing the obtained user habit-using feature vector with the environmental standard pattern set in the functional compartment, specifically: in the same environmental mode, the corresponding items (indoor temperature, humidity, illuminance and CO) are compared2Concentration) is out of regulation.
Selecting training samples based on parameters such as outdoor weather, energy use habits, personnel arrangement and the like, and constructing a support vector machine prediction model;
firstly, for the day to be predicted, a t-period energy consumption characteristic vector x (t) ([ x1, x2, … …, x 12) is constructed]X 1-x 6 respectively correspond to the outdoor current day weather type value, the outdoor current day maximum temperature, the outdoor current day minimum temperature, the outdoor t-time relative humidity and the outdoor t-time solar total radiation, and x 7-x 10 respectively correspond to the indoor t-time temperature corresponding to the energy habit feature vector for the user in the functional areaIndoor t-time relative humidity, indoor t-time illuminance, indoor t-time CO2Concentrations, x11, x12 correspond to the expected number of people and environmental patterns at t in the room. Selecting the corresponding day t moment with the minimum weighted Euclidean distance d as the optimal similar moment from the sample set, wherein
xjiAn ith component representing an energy consumption characteristic vector at the time t on the jth day; Δ xiFor the range of variation of the ith component in the sample set, i.e.
ωiFor weighting, the sperman correlation coefficient of the ith component with the load is adopted.
The constructed support vector machine model has input variables including load values at the optimal similar moment corresponding to each feature vector concentrated by a user's energy habit, the number of indoor persons, an environment mode, outdoor air temperature, outdoor relative humidity, outdoor solar total radiation, outdoor current day weather type values, outdoor current day maximum air temperature, outdoor current day minimum air temperature, load values at 2 moments before the moment, the number of indoor persons, the environment mode, the outdoor air temperature, the outdoor relative humidity, the outdoor solar total radiation at the moment, and the daily weather type values, the outdoor maximum air temperature and the outdoor minimum air temperature. The output is the load value at the predicted time.
The user in step four uses the set of habits for selection of the best similar moment in the sample and modeling of the support vector machine. The user energy use habit set comprises one or more user energy use habit feature vectors, and the energy consumption feature vectors x 7-x 10 at the optimal similar moment are selected to correspond to the user energy use habit feature vectors; in addition, the number of the input of the support vector machine model is also influenced by the number of the feature vectors which can be used by users to concentrate.
The step of constructing the support vector machine model for load prediction comprises the following steps:
(1) selecting N moments with the minimum weighted Euclidean distance d as moments to be predicted to construct a training sample set according to the energy consumption characteristic vector of the predicted moment;
(2) normalizing the training samples to eliminate the influence of different dimensions;
(3) optimizing the model parameters of the support vector machine based on a PSO-LSSVM algorithm to obtain a support vector machine prediction model;
the prediction model input-output relationship is expressed as:
wherein, K (x)i,xj) For the kernel function, an RBF kernel function is adopted:
parameter alphaiB can be calculated by:
wherein (x)i,yi) For the training samples, x is the prediction input vector,
is a predicted value; alpha is alphaiLagrange multipliers, b is a bias parameter; gamma is a penalty coefficient and sigma is a kernel width parameter.
And optimizing the model parameters (gamma, sigma) by adopting a PSO algorithm to minimize the root mean square error of prediction, and obtaining a corresponding support vector machine prediction model.
(4) Inputting weather, environment modes and personnel arrangement information of a specified time period of a forecast day into a forecast model to obtain a load forecast result;
and step six, rolling and predicting cold and heat and electric loads time by time aiming at each functional area to obtain an energy utilization load curve of a prediction day.
And step seven, calculating a target gap according to the indoor environment standard pattern set and the predicted load value, and performing online monitoring and early warning.
Specifically, indoor environmental parameters and load information are monitored in real time, an energy consumption abnormity early warning is sent out for the situations that the environmental parameter adjustment is out of limit, the load deviation value is larger than a preset threshold value and the like by comparing an indoor environmental standard mode set and a predicted energy consumption load curve.
The technical scheme of the invention can carry out on-line real-time detection and diagnosis on the energy consumption condition, the utilization efficiency, the environmental index and the like of the building, eliminate the energy-saving barrier and the waste reason and provide problem reminding or improvement measures.
The online monitoring based on the indoor environment standard mode set and the predicted load value mainly comprises two parts:
firstly, parameters exceeding the upper and lower limits of the corresponding indoor environment standard mode adjustment can be used as energy-saving barriers, for example, the lowest indoor set temperature in summer is 24 degrees, and the actual indoor air conditioner refrigeration reached temperature is lower than 24 degrees, so that the parameters can be used as waste items; or when the room is unattended for a period of time and the power is continuously supplied.
Secondly, the actual operation load of a certain partition is larger than the predicted load, the deviation is larger than a set value, the deviation of the actual operation condition and the original predicted environment condition is analyzed according to the environment condition, and whether energy-saving obstacles exist is determined.
According to the technical scheme, the influence of indoor personnel, energy utilization preference, environmental mode standard and other factors on load prediction is considered, on one hand, the diversity of user habits is fully considered and taken care of, and the user experience is improved; on the other hand, energy consumption waste links are discriminated and corrected, and energy utilization efficiency and load prediction accuracy are improved. For example, when no person is in a certain area during the night, the lighting and the air conditioner are both working, the energy consumption load demand at the moment is actually small, but the load is large in terms of energy consumption data, if the actual demand of the situation cannot be distinguished from the historical energy consumption load data alone, the prediction result is large, therefore, comprehensive identification needs to be carried out by combining an indoor environment model, indoor personnel situations and the like, and the energy consumption behavior habit of a user is identified through clustering, so that the prediction result can reflect the actual demand better.
Example two
It is an object of this embodiment to provide a computing device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the above method when executing the program.
EXAMPLE III
An object of the present embodiment is to provide a computer-readable storage medium.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method.
Example four
The purpose of this embodiment is to provide a building integrated energy load prediction system, includes:
an indoor environment standard pattern set establishment module configured to: dividing a building into different functional areas, and establishing an indoor environment standard mode set aiming at each different functional area;
each functional area energy use habit set establishing module is configured to: clustering environmental data and personnel information in each functional area, and performing benchmarking correction on the environmental data and the personnel information with a standard mode set to form an energy use habit set of each functional area;
a support vector machine prediction model construction module configured to: constructing a support vector machine prediction model based on outdoor meteorological environment data, energy utilization habits and personnel arrangement information parameters;
and obtaining a load prediction result of a prediction day based on the support vector machine prediction model.
In an embodiment, the method specifically further includes: a real-time monitoring module configured to: and monitoring indoor environment parameters and load information in real time, contrasting an indoor environment standard pattern set and a load prediction result, and sending out energy consumption abnormity early warning when the environmental parameter adjustment is over-limit and the load deviation value is greater than a preset threshold value.
The steps involved in the apparatuses of the above second, third and fourth embodiments correspond to the first embodiment of the method, and the detailed description thereof can be found in the relevant description of the first embodiment. The term "computer-readable storage medium" should be taken to include a single medium or multiple media containing one or more sets of instructions; it should also be understood to include any medium that is capable of storing, encoding or carrying a set of instructions for execution by a processor and that cause the processor to perform any of the methods of the present invention.
Those skilled in the art will appreciate that the modules or steps of the present invention described above can be implemented using general purpose computer means, or alternatively, they can be implemented using program code that is executable by computing means, such that they are stored in memory means for execution by the computing means, or they are separately fabricated into individual integrated circuit modules, or multiple modules or steps of them are fabricated into a single integrated circuit module. The present invention is not limited to any specific combination of hardware and software.
Although the embodiments of the present invention have been described with reference to the accompanying drawings, it is not intended to limit the scope of the present invention, and it should be understood by those skilled in the art that various modifications and variations can be made without inventive efforts by those skilled in the art based on the technical solution of the present invention.
Claims (10)
1. A building comprehensive energy load prediction method is characterized by comprising the following steps:
dividing a building into different functional areas;
establishing an indoor environment standard mode set aiming at different functional areas;
clustering environmental data and personnel information in each functional area, and performing benchmarking correction on the environmental data and the personnel information with a standard mode set to form an energy use habit set of each functional area;
constructing a support vector machine prediction model based on outdoor meteorological environment data, energy utilization habits and personnel arrangement information parameters;
and obtaining a load prediction result of a prediction day based on the support vector machine prediction model.
2. The method as claimed in claim 1, further comprising: and monitoring indoor environment parameters and load information in real time, comparing the indoor environment standard mode set with the load prediction result, and sending out energy consumption abnormity early warning when the environmental parameter adjustment is over-limit and the load deviation value is greater than a preset threshold value.
3. The method as claimed in claim 1, wherein the indoor environment standard pattern set includes possible environment patterns of the functional area 24 hours a day, indoor temperature, humidity, illuminance and CO corresponding to each pattern2The expected concentration value, the upper and lower adjustment limits thereof and the corresponding running time of each mode.
4. The method as claimed in claim 1, further comprising obtaining indoor environmental data and personnel information, outdoor weather environmental data, and energy load data for cooling, heating and power;
preferably, the indoor environment data and the personal information include indoor time-by-time air temperature, indoor time-by-time relative humidity, indoor time-by-time illuminance, and indoor time-by-time CO2Concentration, indoor time-by-time personnel number and corresponding environment modes;
the outdoor meteorological environment data comprise outdoor weather types, outdoor current day highest temperature, outdoor current day lowest temperature, outdoor time-by-time relative humidity and outdoor time-by-time solar total radiation;
the energy consumption load data includes time-by-time cooling/heating load and time-by-time power consumption of the functional area.
5. The method as claimed in claim 4, wherein the indoor and outdoor environmental parameters and the time-by-time load data are obtained by calculation using an accumulative average method based on the second-level historical data; the cold/heat load at a certain moment is obtained by calculating the water supply flow of the air conditioner and the temperature difference of the supplied and returned water.
6. The method for predicting the comprehensive energy load of the building as claimed in claim 1, wherein the way of forming the energy use habit set of each functional area is as follows:
clustering indoor environment parameter historical data in the same energy supply cycle and similar time periods in the same period of the previous year in the same mode time by time aiming at the given time period of each functional area to obtain a user energy consumption habit feature vector;
and then, comparing the obtained user energy use habit feature vector with an environment standard pattern set in the functional area, performing limitation correction on parameter components exceeding the upper and lower adjustment limits, and combining to form the user energy use habit set of the functional area.
7. The method for predicting the load of the building integrated energy resource as claimed in claim 1, wherein the step of constructing the support vector machine prediction model comprises the following steps:
selecting N moments with the minimum weighted Euclidean distance as moments to be predicted to construct a training sample set according to the energy consumption characteristic vector of the predicted moment;
normalizing the training samples to eliminate the influence of different dimensions;
optimizing the model parameters of the support vector machine to obtain a prediction model of the support vector machine;
preferably, the energy consumption feature vector at the prediction time is specifically constructed as follows:
constructing a t-period energy consumption characteristic vector X (t) ([ x1, x2, … … and x 12) for the day to be predicted]X 1-x 6 respectively correspond to the outdoor current day weather type value, the outdoor current day maximum temperature, the outdoor current day minimum temperature, the outdoor t-time relative humidity and the outdoor t-time solar total radiation, and x 7-x 10 respectively correspond to the indoor t-time temperature, the indoor t-time relative humidity, the outdoor t-time solar total radiation and the like corresponding to the habituable feature vector for the user in the functional area,Indoor illuminance at t hour and indoor CO at t hour2Concentration, x11, x12 corresponds to the expected number of people and environmental pattern at indoor t;
preferably, at the time of prediction, the cooling/heating and the electric load are predicted for each functional region by time-wise rolling, and an energy consumption load curve of the prediction day is obtained.
8. A building comprehensive energy load prediction system is characterized by comprising:
an indoor environment standard pattern set establishment module configured to: dividing a building into different functional areas, and establishing an indoor environment standard mode set aiming at each different functional area;
each functional area energy use habit set establishing module is configured to: clustering environmental data and personnel information in each functional area, and performing benchmarking correction on the environmental data and the personnel information with a standard mode set to form an energy use habit set of each functional area;
a support vector machine prediction model construction module configured to: constructing a support vector machine prediction model based on outdoor meteorological environment data, energy utilization habits and personnel arrangement information parameters;
obtaining a load prediction result of a prediction day based on a support vector machine prediction model;
preferably, the method further comprises the following steps: a real-time monitoring module configured to: and monitoring indoor environment parameters and load information in real time, contrasting an indoor environment standard pattern set and a load prediction result, and sending out energy consumption abnormity early warning when the environmental parameter adjustment is over-limit and the load deviation value is greater than a preset threshold value.
9. A computing device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program performs the steps of the method of any one of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, is adapted to carry out the steps of the method according to any one of the preceding claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110990133.8A CN113902582A (en) | 2021-08-26 | 2021-08-26 | Building comprehensive energy load prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110990133.8A CN113902582A (en) | 2021-08-26 | 2021-08-26 | Building comprehensive energy load prediction method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113902582A true CN113902582A (en) | 2022-01-07 |
Family
ID=79188014
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110990133.8A Pending CN113902582A (en) | 2021-08-26 | 2021-08-26 | Building comprehensive energy load prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113902582A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114777306A (en) * | 2022-04-15 | 2022-07-22 | 北京德克浦智慧能源科技有限公司 | Intelligent air conditioner regulation and control method and system, electronic equipment and storage medium |
| CN114880754A (en) * | 2022-07-07 | 2022-08-09 | 青岛黄海学院 | BIM-based building energy consumption management method and system |
| CN115186892A (en) * | 2022-07-08 | 2022-10-14 | 建德市安泰科技咨询有限公司 | Energy-saving assessment method, device and system based on big data |
| CN115796390A (en) * | 2023-01-31 | 2023-03-14 | 北京智能建筑科技有限公司 | Data-driven building energy consumption abnormity diagnosis method and system |
| CN116542395A (en) * | 2023-06-12 | 2023-08-04 | 重庆不贰科技(集团)有限公司 | Low-carbon building monitoring system and method |
-
2021
- 2021-08-26 CN CN202110990133.8A patent/CN113902582A/en active Pending
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114777306A (en) * | 2022-04-15 | 2022-07-22 | 北京德克浦智慧能源科技有限公司 | Intelligent air conditioner regulation and control method and system, electronic equipment and storage medium |
| CN114777306B (en) * | 2022-04-15 | 2024-05-17 | 北京德克浦智慧能源科技有限公司 | Intelligent air conditioner regulation and control method and system, electronic equipment and storage medium |
| CN114880754A (en) * | 2022-07-07 | 2022-08-09 | 青岛黄海学院 | BIM-based building energy consumption management method and system |
| CN114880754B (en) * | 2022-07-07 | 2022-09-23 | 青岛黄海学院 | BIM-based building energy consumption management method and system |
| CN115186892A (en) * | 2022-07-08 | 2022-10-14 | 建德市安泰科技咨询有限公司 | Energy-saving assessment method, device and system based on big data |
| CN115186892B (en) * | 2022-07-08 | 2023-09-15 | 建德市安泰科技咨询有限公司 | Big data-based energy-saving evaluation method, device and system |
| CN115796390A (en) * | 2023-01-31 | 2023-03-14 | 北京智能建筑科技有限公司 | Data-driven building energy consumption abnormity diagnosis method and system |
| CN116542395A (en) * | 2023-06-12 | 2023-08-04 | 重庆不贰科技(集团)有限公司 | Low-carbon building monitoring system and method |
| CN116542395B (en) * | 2023-06-12 | 2024-01-26 | 重庆不贰科技(集团)有限公司 | Low-carbon building monitoring system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Alanne et al. | An overview of machine learning applications for smart buildings | |
| CN113902582A (en) | Building comprehensive energy load prediction method and system | |
| CN113962364B (en) | Multi-factor power load prediction method based on deep learning | |
| Brown et al. | Kernel regression for real-time building energy analysis | |
| CN106951611A (en) | A kind of severe cold area energy-saving design in construction optimization method based on user's behavior | |
| CN113112077B (en) | HVAC control system based on multi-step prediction deep reinforcement learning algorithm | |
| CN115249094B (en) | Building energy efficiency management and optimization method based on big data | |
| CN112232575B (en) | Comprehensive energy system regulation and control method and device based on multi-element load prediction | |
| CN110796307A (en) | Distributed load prediction method and system for comprehensive energy system | |
| CN112598195A (en) | Building type comprehensive energy system operation optimization method and device and terminal equipment | |
| Higuera et al. | Hybrid smart lighting and climate control system for buildings | |
| Song et al. | An indoor temperature prediction framework based on hierarchical attention gated recurrent unit model for energy efficient buildings | |
| Wang et al. | Multi-criteria comprehensive study on predictive algorithm of heating energy consumption of district heating station based on timeseries processing | |
| CN112987617A (en) | Near-zero energy consumption building digital management system and energy efficiency monitoring method | |
| Qin et al. | Energy-efficient heating control for nearly zero energy residential buildings with deep reinforcement learning | |
| CN113887809A (en) | Power distribution network supply and demand balance method, system, medium and computing equipment under double-carbon target | |
| Wang | Application of fuzzy linear programming model in agricultural economic management | |
| CN112307675A (en) | Neural network-based temperature-sensitive load separation identification method and system | |
| Liu et al. | Social implications of cyber-physical systems in electrical load forecasting | |
| CN113609778B (en) | Multi-objective optimization method and system for comprehensive energy system | |
| Qiao | Intelligent building with multi-energy system planning method considering energy supply reliability | |
| CN115309053A (en) | Data-driven multi-element fusion high-elasticity power grid resource cooperative regulation and control system | |
| CN115169839A (en) | Heating load scheduling method based on data-physics-knowledge combined drive | |
| Liu et al. | Occupant-centric HVAC and window control: A reinforcement learning model for enhancing indoor thermal comfort and energy efficiency | |
| Liu et al. | Enhancing HVAC energy management through multi-zone occupant-centric approach: A multi-agent deep reinforcement learning solution |
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 |