CN115628522A - Market central air-conditioning load prediction method, system and medium based on EMD-PSO-LSTM - Google Patents
Market central air-conditioning load prediction method, system and medium based on EMD-PSO-LSTM Download PDFInfo
- Publication number
- CN115628522A CN115628522A CN202211343633.3A CN202211343633A CN115628522A CN 115628522 A CN115628522 A CN 115628522A CN 202211343633 A CN202211343633 A CN 202211343633A CN 115628522 A CN115628522 A CN 115628522A
- Authority
- CN
- China
- Prior art keywords
- load
- lstm
- emd
- central air
- pso
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- 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
- F24F11/64—Electronic processing using pre-stored data
-
- 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
- F24F11/89—Arrangement or mounting of control or safety devices
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (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)
- Air Conditioning Control Device (AREA)
Abstract
The invention discloses a method, a system and a medium for predicting the load of a central air conditioner in a market based on EMD-PSO-LSTM. The method comprises the steps of collecting temperature and flow data of a chilled water system of the central air conditioner and outdoor temperature and humidity data, and carrying out load calculation; constructing an LSTM model, training based on an EMD decomposition method, and performing parameter optimization by using a PSO method; and predicting the load according to the historical load and the outdoor temperature and humidity data. According to the invention, EMD can convert a non-stationary sequence into a stationary sequence, so that the non-linearity is reduced, and the prediction precision and speed are improved. In addition, the invention also utilizes historical data to dig out the relevance among the loads, thereby improving the prediction precision of the model.
Description
Technical Field
The invention belongs to the technical field of building energy conservation, and particularly relates to a method, a system and a medium for predicting the load of a market central air conditioner based on EMD-PSO-LSTM.
Background
The energy-saving promotion engineering of the urban building requires green and efficient refrigeration action, the refrigeration technology and equipment are updated and upgraded, the matching of load supply and demand is optimized, and the energy efficiency level of a refrigeration system is greatly promoted. The large-scale comprehensive market is used as an important component of a large-scale public building, has the characteristics of large building area, large window-wall ratio, high personnel density, long operation time, high density of various lighting appliances, high energy consumption of a central air conditioner and the like, has energy consumption per unit area far higher than that of other large-scale public buildings, and has huge energy-saving potential.
In the design stage of the market central air-conditioning system, designers are all designed according to the maximum cooling load in order to ensure that the cooling demand for comfort in the market can be met in extreme weather or the maximum passenger flow. In the actual operation stage, the cold load of the shopping mall is influenced by a plurality of factors such as outdoor meteorological parameters, indoor passenger flow density and the like to dynamically change, and the central air-conditioning system is in a partial load operation state 90% of the time. Due to the lack of corresponding management measures, cold quantity adjusting means and control devices, the market central air-conditioning system cannot perform dynamic optimization adjustment according to cold load, cold quantity is over supplied and over requested, the operation energy efficiency is low, and a large amount of energy is wasted. Therefore, accurate prediction of the central air-conditioning load in the market is the basis for realizing the dynamic adjustment of the air-conditioning system.
Disclosure of Invention
The invention mainly aims to overcome the defects in the prior art, provides a method, a system and a medium for predicting the load of a mall central air conditioner based on EMD-PSO-LSTM, and can realize accurate prediction of the load by combining historical data and outdoor real-time meteorological parameters. Compared with the existing market central air-conditioning load prediction method, the prediction precision can be further improved.
In order to achieve the purpose, the invention adopts the following technical scheme:
one aspect of the invention provides a market central air-conditioning load prediction method based on EMD-PSO-LSTM, which comprises the following steps:
collecting temperature and flow data of a chilled water system of a central air conditioner and outdoor temperature and humidity data, and carrying out load calculation;
constructing an LSTM model, training based on an EMD decomposition method, and performing parameter optimization by using a PSO method;
and predicting the load according to the historical load and the outdoor temperature and humidity data.
The preferable technical scheme is characterized in that the collecting of the temperature and flow data of the chilled water system of the central air conditioner specifically comprises the following steps:
arranging a temperature sensor in a chilled water system of the central air conditioner, and setting the temperature sensor at the water separator as the supply water temperature T 1 Setting the temperature sensor at the water collector to return water temperature T 2 ；
And arranging a flow sensor in the chilled water system of the central air conditioner, and setting the flow at the chilled water main pipe as the chilled water flow q.
As a preferred technical solution, the load calculation is specifically:
Q＝q×(T 2 -T 1 )×c×ρ
wherein Q is the load at the current moment, c is the specific heat capacity of the chilled water, and rho is the density of the chilled water.
The preferable technical scheme is characterized in that the construction of the LSTM model specifically comprises the following steps:
f t ＝σ(W f ·[h t-1 ,x t ]+b f )
i t ＝σ(W i ·[h t-1 ,x t ]+b i )
O t ＝σ(W O ·[h t-1 ,x t ]+b O )
g t ＝O t *tan h(C t )
wherein, W is a weight term, b is a bias term and is a sigmoid function; input at each momentThe variable contains the state C of the cell at the previous time t-1 Intermediate state h at the previous moment t-1 And input x at the current time t The intermediate variable comprising the output f of the forgetting gate t Output of input-output gate i t And O t And the output of the input node
The output variables include cell state C t And an intermediate state h t 。
The preferable technical scheme is characterized in that the training of the LSTM model specifically comprises the following steps:
decomposing the load sequence x (t) by using EMD to obtain a plurality of IMFs and a residual error;
splicing the load, the outdoor temperature and the outdoor humidity at the first 2 moment and the first 1 moment into a data sequence according to a, b and c;
converting the data sequence into a format of nx3x3, inputting the data sequence into an LSTM model, and training by taking the load Q at the current moment as a predicted value;
optimizing the learning rate, the number of hidden layers, the number of nodes in each layer and the iteration number of the LSTM model by adopting a particle swarm algorithm;
predicting each IMF according to the steps to obtain predicted values from 0 to n, wherein n is the number of IMFs;
superposing the predicted values from 0 to n and the residual error to obtain a load predicted value;
and saving the step model.
As a preferred technical solution, the load sequence x (t) is decomposed by using EMD, specifically:
finding all extreme points in the load sequence x (t);
connecting all maximum points by using envelope lines to form e max Similarly, all the minimum value points are connected to form e min ；
The average e of the upper and lower envelopes is determined mean And subtracting it from the original sequence to give a new sequence H:
H＝x(t)-e mean
judging whether the new sequence H is an intrinsic mode function IFM according to the following criteria:
a) The difference between the number of extreme points and the number of zero points in the intrinsic mode function IFM is not more than 1;
b) The mean value of the upper envelope line and the lower envelope line of the intrinsic mode function IFM at any moment is 0;
c) If not, taking H as new x (t), and repeating the steps until the criterion is met to obtain IFM0;
each time the eigenmode function IFM is obtained, it is removed from x (t):
x(t)＝x(t)-IFM0
repeating the steps until the residual error Res of the residual part is a monotone sequence or a constant value sequence, namely:
x(t)＝IFM0+IMF1+IMF2+…+IMFn+Res。
as a preferred technical solution, the particle swarm algorithm specifically comprises:
randomly initializing each particle;
evaluating each particle and obtaining a global optimum;
judging whether an ending condition is met, if so, ending;
if the ending condition is not met, updating the position and the speed of each particle;
updating the fitness function of each particle;
updating the historical optimal position of each particle;
returning to the step of evaluating each particle and obtaining the global optimum;
and judging whether the ending condition is met.
As a preferred technical scheme, the load prediction according to the historical load and the outdoor temperature and humidity data specifically includes:
calculating to obtain the actual load Q at the first 1 moment according to the data collected by the history t-1 And actual load Q at the first 2 time t-2 To do so byAnd outdoor temperature T at the first 2 nd moment 3,t-2 Outdoor humidity w at the first 2 moments t-2 Outdoor temperature T at the previous 1 st moment 3,t-1 Outdoor humidity w at the first 1 moment t-1 Inputting the load as an input item into an LSTM model for prediction to obtain a load predicted value Q at the current time t 。
The invention provides a market central air-conditioning load prediction system based on EMD-PSO-LSTM, which is applied to the market central air-conditioning load prediction method based on EMD-PSO-LSTM and comprises a data acquisition and calculation module, a model construction training module and a prediction module;
the data acquisition and calculation module is used for acquiring temperature and flow data of a chilled water system of the central air conditioner and outdoor temperature and humidity data and carrying out load calculation;
the model construction training module is used for constructing an LSTM model, training the LSTM model based on an EMD decomposition method and optimizing parameters by using a PSO method;
the prediction module is used for predicting the load according to the historical load and the outdoor temperature and humidity data.
In another aspect of the present invention, a storage medium is provided, which stores a program, and when the program is executed by a processor, the program implements the method for predicting the load of the central air conditioner in the market based on EMD-PSO-LSTM.
Compared with the prior art, the invention has the following advantages and beneficial effects:
(1) The EMD can convert a non-stationary sequence into a stationary sequence, so that the nonlinearity is reduced, and the prediction precision and speed are improved.
(2) The method for rapidly realizing the load prediction of the central air conditioner in the shopping mall can shorten the prediction time.
(3) And mining the relevance among the loads by using historical data, and improving the model prediction precision.
Drawings
FIG. 1 is a flow chart of a method for predicting the load of a central air conditioner in a market based on EMD-PSO-LSTM according to an embodiment of the present invention;
FIG. 2 is a schematic structural diagram of a central air conditioning system of a mall in accordance with an embodiment of the present invention;
FIG. 3 is a schematic diagram of data sequence generation according to an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of an LSTM according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of an LSTM prediction structure according to an embodiment of the present invention;
FIG. 6 is a flow chart of a PSO algorithm according to an embodiment of the present invention;
FIG. 7 is a schematic view of the central air-conditioning load prediction in the mall of the embodiment of the present invention;
FIG. 8 is a schematic structural diagram of a central air conditioning load forecasting system in a market based on EMD-PSO-LSTM according to an embodiment of the present invention;
fig. 9 is a schematic structural diagram of a storage medium according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. It should be apparent that the described embodiments are only a few embodiments of the present application, and not all embodiments. All other embodiments obtained by a person skilled in the art based on the embodiments in the present application without making any creative effort belong to the protection scope of the present application.
Examples
In this embodiment, a data acquisition system is used to construct a central air conditioning load sequence in a mall, and an EMD (Empirical Mode Decomposition) is used to decompose the load sequence into a plurality of IMF (Intrinsic Mode Function) sequences and residuals, thereby reducing non-linearity. And forming an input sequence by the IMF and the outdoor meteorological parameters, and inputting the input sequence into an LSTM (Long Short-Term Memory neural network) for load prediction. Meanwhile, PSO (Particle Swarm Optimization) is adopted to optimize LSTM related hyper-parameters, and the prediction speed is accelerated. And finally, overlapping the outputs to obtain a load predicted value.
As shown in fig. 1, the embodiment provides a method for predicting the load of a central air conditioner in a market based on EMD-PSO-LSTM, which includes the following steps:
s1, data acquisition and calculation.
Arranging a temperature sensor in a chilled water system of the central air conditioner, and setting the temperature sensor at the water separator as the supply water temperature T 1 Setting the temperature sensor at the water collector to return water temperature T 2 (see fig. 1).
A flow sensor is arranged in the chilled water system of the central air conditioner, and the flow at the chilled water main is set as the chilled water flow q (as shown in fig. 2).
Arranging a temperature and humidity sensor outdoors to set the outdoor temperature to T 3 And the outdoor humidity is set to w.
Data were collected every 20 minutes and according to formula Q = qx (T) 2 -T 1 ) And multiplying by x c rho, calculating the load Q at the current moment (c is the specific heat capacity of the chilled water, and rho is the density of the chilled water).
Calculating the load Q and the outdoor temperature T 3 The outdoor humidity w is spliced into a sequence according to a, b and c (as shown in fig. 3).
And S2, constructing an LSTM model and training.
(1) An LSTM model structure;
as shown in fig. 4, the LSTM model in this embodiment is specifically:
f t ＝σ(W f ·[h t-1 ,x t ]+b f )
i t ＝σ(W i ·[h t-1 ,x t ]+b i )
O t ＝σ(W O ·[h t-1 ,x t ]+b O )
h t ＝O t *tan h(C t )
whereinWherein W is a weight term, b is a bias term and is a sigmoid function; the input variable at each time contains the state C of the cell at the previous time t-1 Intermediate state h at the previous moment t-1 And input x at the current time t The intermediate variable comprising the output f of the forgetting gate t Output of input-output gate i t And O t And the output of the input node
The output variables include cell state C t And an intermediate state h t 。
(2) The LSTM model training process is as follows:
s2.1, decomposing the load sequence x (t) by using EMD to obtain a plurality of IMFs and a residual error.
The load sequence x (t) is a sequence x (t) in which the loads from time t =0 to time t = n collected in step S1 are combined.
Further, in this embodiment, the decomposing the payload sequence x (t) by using the EMD specifically includes:
finding all extreme points in the load sequence x (t);
connecting all maximum points by using envelope lines to form e max Similarly, all the minimum value points are connected to form e min ；
The average e of the upper and lower envelopes is determined mean And subtracting it from the original sequence to give a new sequence H:
H＝x(t)-e mean
judging whether the new sequence H is an intrinsic mode function IFM according to the following criteria:
a) The difference between the number of extreme points and the number of zero points in the intrinsic mode function IFM is not more than 1;
b) The mean value of the upper envelope line and the lower envelope line of the intrinsic mode function IFM at any moment is 0;
c) If not, taking H as new x (t), and repeating the steps until the criterion is met to obtain IFM0;
every time the eigenmode function IFM is obtained, it is removed from x (t):
x(t)＝x(t)-IFM0
repeating the steps until the residual errors Rea of the rest parts are monotonous sequences or constant value sequences, namely:
z(t)＝IFM0+IMF1+IMF2+…+IMFn+Res。
s2.2, the data sequence is converted into a format of nx3x3, and input to the LSTM model, and training is performed with the load Q at the current time as a predicted value (as shown in fig. 5).
S2.3, optimizing the learning rate, the number of hidden layers, the number of nodes in each layer and the iteration number of the LSTM by using a particle swarm algorithm (as shown in figure 6).
Further, as shown in fig. 6, the particle swarm algorithm specifically includes:
randomly initializing each particle;
evaluating each particle and obtaining a global optimum;
judging whether an ending condition is met, if so, ending;
if the ending condition is not met, updating the position and the speed of each particle;
updating the fitness function of each particle;
updating the historical optimal position of each particle;
returning to the step of evaluating each particle and obtaining the global optimum;
and judging whether the ending condition is met.
S2.4, as shown in fig. 7, the prediction is performed for each IMF according to the above steps to obtain predicted values 0 to n.
The inputs to IMF0 are:
| Q t-2,0 | T 3,t-2 | w t-2 |
| Q t-1,0 | T 3,t-1 | w t-1 |
the output is: q t,0 ；
The inputs to IMFn are:
| Q t-2,n | T 3,t-2 | w t-2 |
| Q t-1,n | T 3,t-1 | w t-1 |
the output is: q t,n 。
S2.5, superposing the predicted value 0, the predicted value 1\8230, the predicted value n and the residual error to obtain the load predicted value.
And S2.6, storing the model in the step.
And S3, load prediction (taking the time t as an example).
Calculating to obtain the actual load Q at the first 1 moment from the data collected in the step S1 t-1 And actual load Q at the first 2 time t-2 And outdoor temperature T at the first 2 time 3,t-2 And outdoor humidity w at the first 2 moments t-2 Outdoor temperature T at the previous 1 st moment 3,t-1 Outdoor humidity w at the first 1 moment t-1 Inputting the load as an input item into an LSTM model for prediction to obtain a load predicted value Q at the current time t 。
As shown in fig. 8, in another embodiment of the present application, there is provided an EMD-PSO-LSTM-based central air-conditioning load forecasting system for a shopping mall, the system includes a data acquisition and calculation module, a model construction training module and a forecasting module;
the data acquisition and calculation module is used for acquiring temperature and flow data of a chilled water system of the central air conditioner and outdoor temperature and humidity data and carrying out load calculation;
the model construction training module is used for constructing an LSTM model, training the LSTM model based on an EMD decomposition method and optimizing parameters by using a PSO method;
the prediction module is used for predicting the load according to the historical load and the outdoor temperature and humidity data.
It should be noted that the system provided in the above embodiment is only illustrated by the division of the above functional modules, and in practical applications, the above function allocation may be completed by different functional modules according to needs, that is, the internal structure is divided into different functional modules to complete all or part of the above described functions.
As shown in fig. 9, in another embodiment of the present application, there is further provided a storage medium storing a program, which when executed by a processor, implements the EMD-PSO-LSTM-based mall central air conditioning load prediction of the foregoing embodiment, specifically:
collecting temperature and flow data of a chilled water system of a central air conditioner and outdoor temperature and humidity data, and carrying out load calculation;
constructing an LSTM model, training based on an EMD decomposition method, and performing parameter optimization by using a PSO method;
and predicting the load according to the historical load and the outdoor temperature and humidity data.
It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
The above embodiments are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents thereof, and all such modifications are intended to be included in the scope of the present invention.
Claims (10)
1. The method for predicting the load of the central air conditioner in the market based on the EMD-PSO-LSTM is characterized by comprising the following steps:
collecting temperature and flow data of a chilled water system of a central air conditioner and outdoor temperature and humidity data, and carrying out load calculation;
constructing an LSTM model, training based on an EMD decomposition method, and performing parameter optimization by using a PSO method;
and predicting the load according to the historical load and the outdoor temperature and humidity data.
2. The emporium central air-conditioning load prediction method based on EMD-PSO-LSTM according to claim 1, wherein the data of temperature and flow of the central air-conditioning chilled water system are collected, specifically:
arranging a temperature sensor in a chilled water system of the central air conditioner, and setting the temperature sensor at the water separator as the supply water temperature T 1 Will collectThe temperature sensor at the water heater is set as the return water temperature T 2 ；
And arranging a flow sensor in the chilled water system of the central air conditioner, and setting the flow at the chilled water main pipe as the chilled water flow q.
3. The emporium central air-conditioning load prediction method based on the EMD-PSO-LSTM as claimed in claim 2, wherein the load calculation specifically comprises:
Q＝q×(T 2 -T 1 )×c×ρ
wherein Q is the load at the current moment, c is the specific heat capacity of the chilled water, and rho is the density of the chilled water.
4. The emporium central air-conditioning load prediction method based on the EMD-PSO-LSTM as claimed in claim 1, wherein the constructing of the LSTM model specifically comprises:
f t ＝σ(W f ·[h t-1 ,x t ]+b f )
i t ＝σ(W i ·[h t-1 ,x t ]+b i )
O t ＝σ(W o ·[h t-1 ,x t ]+b O )
h t ＝O t *tanh(C t )
wherein, W is a weight term, b is a bias term and is a sigmoid function; the input variable at each time contains the state C of the cell at the previous time t-1 Intermediate state h at the previous moment t-1 And input x at the current time t The intermediate variable comprising the output f of the forgetting gate t Of input-output gatesOutput i t And O t And the output of the input node
The output variables include cell state C t And an intermediate state h t 。
5. The emporium central air-conditioning load prediction method based on EMD-PSO-LSTM as claimed in claim 1, wherein the LSTM model training specifically comprises:
decomposing the load sequence x (t) by using EMD to obtain a plurality of IMFs and a residual error;
splicing the load, the outdoor temperature and the outdoor humidity at the first 2 moment and the first 1 moment into a data sequence according to a, b and c;
converting the data sequence into a format of nx3x3, inputting the data sequence into an LSTM model, and training by taking the load Q at the current moment as a predicted value;
optimizing the learning rate, the number of hidden layers, the number of nodes in each layer and the iteration number of the LSTM model by adopting a particle swarm algorithm;
predicting each IMF according to the steps to obtain predicted values from 0 to n, wherein n is the number of IMFs;
superposing the predicted values from 0 to n and the residual error to obtain a load predicted value;
and saving the step model.
6. The method for predicting the load of the emporium central air conditioner based on the EMD-PSO-LSTM as claimed in claim 5, wherein the EMD is used for decomposing the load sequence x (t), and specifically comprises the following steps:
finding all extreme points in the load sequence x (t);
connecting all maximum points by using envelope lines to form e max Similarly, all the minimum value points are connected to form e min ；
The average e of the upper and lower envelopes is determined mean And subtracting it from the original sequence to give a new sequence H:
H＝x(t)-e mean
judging whether the new sequence H is an intrinsic mode function IFM according to the following criteria:
a) The difference between the number of extreme points and the number of zero points in the intrinsic mode function IFM is not more than 1;
b) The mean value of the upper envelope line and the lower envelope line of the intrinsic mode function IFM is 0 at any time;
c) If not, taking H as new x (t), and repeating the steps until the criterion is met to obtain IFM0;
every time the eigenmode function IFM is obtained, it is removed from x (t):
x(t)＝x(t)-IFM0
repeating the steps until the residual error Res of the residual part is a monotone sequence or a constant value sequence, namely:
x(t)＝IFM0+IMF1+IMF2+…+IMFn+Res。
7. the method for predicting the load of the emporium central air conditioner based on the EMD-PSO-LSTM as claimed in claim 5, wherein the particle swarm algorithm is specifically as follows:
randomly initializing each particle;
evaluating each particle and obtaining a global optimum;
judging whether an ending condition is met, if so, ending;
if the ending condition is not met, updating the position and the speed of each particle;
updating the fitness function of each particle;
updating the historical optimal position of each particle;
returning to the step of evaluating each particle and obtaining the global optimum;
and judging whether the ending condition is met.
8. The emporium central air-conditioning load prediction method based on the EMD-PSO-LSTM as claimed in claim 1, wherein the load prediction is performed according to historical load and outdoor temperature and humidity data, specifically:
calculating to obtain the actual load Q at the previous 1 moment according to the data collected by history t-1 And actual load Q at the first 2 time t-2 And outdoor temperature T at the first 2 time 3,t-2 Outdoor humidity w at the first 2 moments t-2 Outdoor temperature T at the previous 1 st moment 3,t-1 Outdoor humidity w at the first 1 moment t-1 Inputting the load as an input item into an LSTM model for prediction to obtain a load predicted value Q at the current time t 。
9. A market central air-conditioning load prediction system based on EMD-PSO-LSTM is characterized by being applied to the market central air-conditioning load prediction method based on EMD-PSO-LSTM in any one of claims 1-8, and comprising a data acquisition and calculation module, a model construction training module and a prediction module;
the data acquisition and calculation module is used for acquiring temperature and flow data of a chilled water system of the central air conditioner and outdoor temperature and humidity data and carrying out load calculation;
the model construction training module is used for constructing an LSTM model, training the LSTM model based on an EMD decomposition method and optimizing parameters by using a PSO method;
the prediction module is used for predicting the load according to the historical load and the outdoor temperature and humidity data.
10. A storage medium storing a program, characterized in that: when the program is executed by a processor, the method for predicting the load of the central air conditioner in the market based on the EMD-PSO-LSTM as claimed in any one of claims 1 to 8 is realized.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211343633.3A CN115628522A (en) | 2022-10-31 | 2022-10-31 | Market central air-conditioning load prediction method, system and medium based on EMD-PSO-LSTM |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211343633.3A CN115628522A (en) | 2022-10-31 | 2022-10-31 | Market central air-conditioning load prediction method, system and medium based on EMD-PSO-LSTM |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115628522A true CN115628522A (en) | 2023-01-20 |
Family
ID=84909249
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211343633.3A Pending CN115628522A (en) | 2022-10-31 | 2022-10-31 | Market central air-conditioning load prediction method, system and medium based on EMD-PSO-LSTM |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115628522A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117167903A (en) * | 2023-11-03 | 2023-12-05 | 江苏中安建设集团有限公司 | Artificial intelligence-based foreign matter fault detection method for heating ventilation equipment |
-
2022
- 2022-10-31 CN CN202211343633.3A patent/CN115628522A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117167903A (en) * | 2023-11-03 | 2023-12-05 | 江苏中安建设集团有限公司 | Artificial intelligence-based foreign matter fault detection method for heating ventilation equipment |
| CN117167903B (en) * | 2023-11-03 | 2024-01-30 | 江苏中安建设集团有限公司 | Artificial intelligence-based foreign matter fault detection method for heating ventilation equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113112077B (en) | HVAC control system based on multi-step prediction deep reinforcement learning algorithm | |
| Welch et al. | Energy dispatch fuzzy controller for a grid-independent photovoltaic system | |
| CN109034587B (en) | Active power distribution system optimal scheduling method for coordinating multiple controllable units | |
| CN112070262B (en) | Air conditioner load prediction method based on support vector machine | |
| CN115628522A (en) | Market central air-conditioning load prediction method, system and medium based on EMD-PSO-LSTM | |
| CN114418174A (en) | Electric vehicle charging load prediction method | |
| CN112508279A (en) | Regional distributed photovoltaic prediction method and system based on spatial correlation | |
| Huang et al. | Smart energy management system based on reconfigurable AI chip and electrical vehicles | |
| CN116187173A (en) | Double-layer optimization method and system for regional comprehensive energy system | |
| CN110867902B (en) | Micro-grid distributed power supply decentration optimization operation method based on power generation prediction | |
| CN112446552A (en) | Multi-objective optimization method of biomass gasification combined cooling heating and power system | |
| CN115759604B (en) | Comprehensive energy system optimal scheduling method | |
| Testasecca et al. | Towards Digital Twins of buildings and smart energy networks: Current and future trends | |
| CN116468291A (en) | Commercial building hybrid energy scheduling method with electric vehicle charging station | |
| CN106786800B (en) | Power system economic dispatching analysis method based on big data processing and containing photovoltaic power generation | |
| CN112668764B (en) | Offshore wind farm energy storage system optimization configuration method based on cloud model and FCM algorithm | |
| CN112734451B (en) | Green house multi-energy system based on non-cooperative game and optimization method | |
| Fang et al. | Energy scheduling and decision learning of combined cooling, heating and power microgrid based on deep deterministic policy gradient | |
| CN113949082A (en) | Day-ahead energy scheduling method and device for industrial park | |
| Chinnadurrai et al. | Energy Management of a Microgrid based on LSTM Deep Learning Prediction Model | |
| CN113222214B (en) | Random scene-based optimization design method and system for comprehensive energy system | |
| CN115840986B (en) | Energy management method based on stochastic model predictive control | |
| Zhang et al. | Integration of Urban Energy Systems with Renewable Envelope Solutions at Building Cluster Level | |
| CN114862004A (en) | Energy system multi-energy collaborative prediction method based on neural network | |
| CN118249408A (en) | Grid-connected hybrid renewable energy system based on combination optimization and machine learning algorithm |
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 |