CN116070881B - Intelligent energy consumption scheduling method and device for modern industrial production area - Google Patents
Intelligent energy consumption scheduling method and device for modern industrial production area Download PDFInfo
- Publication number
- CN116070881B CN116070881B CN202310236536.2A CN202310236536A CN116070881B CN 116070881 B CN116070881 B CN 116070881B CN 202310236536 A CN202310236536 A CN 202310236536A CN 116070881 B CN116070881 B CN 116070881B
- Authority
- CN
- China
- Prior art keywords
- data
- energy consumption
- staff
- scheduling
- production line
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000005265 energy consumption Methods 0.000 title claims abstract description 102
- 238000000034 method Methods 0.000 title claims abstract description 19
- 238000009776 industrial production Methods 0.000 title claims abstract description 17
- 238000004519 manufacturing process Methods 0.000 claims abstract description 86
- 230000005611 electricity Effects 0.000 claims abstract description 67
- 238000013461 design Methods 0.000 claims abstract description 45
- 238000013528 artificial neural network Methods 0.000 claims abstract description 10
- 238000003062 neural network model Methods 0.000 claims description 21
- 230000001105 regulatory effect Effects 0.000 claims description 13
- 238000012549 training Methods 0.000 claims description 12
- 241000282346 Meles meles Species 0.000 claims description 11
- 210000002569 neuron Anatomy 0.000 claims description 10
- 238000011161 development Methods 0.000 claims description 9
- 230000009191 jumping Effects 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 9
- 230000001276 controlling effect Effects 0.000 claims description 8
- 230000001133 acceleration Effects 0.000 claims description 6
- 235000012907 honey Nutrition 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 5
- 241000282345 Meles Species 0.000 claims description 3
- 235000013305 food Nutrition 0.000 claims description 3
- 238000011160 research Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06315—Needs-based resource requirements planning or analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Tourism & Hospitality (AREA)
- Software Systems (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Mathematical Physics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Educational Administration (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Power Engineering (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses an intelligent energy consumption scheduling method and device for a modern industrial production area, and the method and device comprise a data acquisition unit, an energy consumption scheduling strategy design unit, an energy consumption scheduling unit and a scheduling object. The data acquisition unit comprises an ammeter and an employee work system collection unit, and collects the electric load demand of the whole industrial park, the electricity consumption of each electric equipment, the working time of the employee and the employee number; the energy consumption scheduling strategy design unit comprises a prediction module and a strategy design module, wherein the energy consumption scheduling strategy design module firstly adopts an LSTM neural network to predict the electricity consumption based on the data acquired by the data acquisition unit, comprises the participation of staff in the electricity consumption prediction of the production line and the full-automatic production line, and then adopts an AOA algorithm to carry out energy consumption scheduling strategy design based on the predicted data; the energy consumption scheduling unit performs energy consumption scheduling of the production area based on a designed scheduling strategy. Compared with the prior art, the application can schedule the energy consumption of the industrial production area, improves the economic benefit of the whole production area and reduces the operation cost on the premise of ensuring the normal operation of the production area.
Description
Technical Field
The application belongs to the technical field of intelligent energy consumption scheduling of equipment, and particularly relates to an intelligent energy consumption scheduling method and device for a modern industrial production area.
Background
Under the global background of increasing energy consumption, increasing energy price and advocating green production, energy conservation and environmental protection, the research value for the energy consumption problem in the production process is gradually increased. Through reasonable energy consumption scheduling, the optimization of economic indexes can be realized. The existing intelligent energy consumption scheduling equipment has the following steps:
first, by establishing an energy consumption optimizing and scheduling model, optimizing and controlling the energy consumption of the manufacturing process while completing the processing task, and taking the energy consumption and the delivery date as scheduling targets in a cost manner, the energy consumption of the manufacturing process can be controlled and optimized while reducing the scheduling cost. The control of the use time of the electric equipment is an important regulation and control index for energy consumption regulation, and the scheme lacks regulation and control scheduling of the use time point of the electric equipment, the working time of staff and the number of people;
second, the power supplier encourages the enterprise to reduce peak electricity consumption by differentiating electricity price setting, avoids the period of high energy price in the single machine energy consumption optimizing model, thereby reducing total power cost, and simultaneously optimizes energy cost (economy) and carbon footprint (ecology) under the time-sharing electricity price environment by the 'ecological scheduling' model, so that the system energy cost is reduced and is more environment-friendly. The energy consumption regulation is also an important regulation index for equipment energy consumption, and the scheme lacks monitoring control for the energy consumption during the valley, which can cause the energy consumption during the valley to be too high and cause accidents.
Two defects exist in the research of workshop energy-saving scheduling, namely, on one hand, the research of equipment use time allocation is insufficient, and the equipment use in valley is not considered to occupy larger power consumption energy; on the other hand, the specific work information of the workshop is not fully collected, and the energy consumption scheduling is performed by not predicting and making a strategy in time. The problems in the two energy consumption scheduling schemes are that the using time point of the electric equipment is not long in time, the working time of workshop staff and the number of people are regulated and scheduled, and the valley time energy consumption of the equipment is not monitored and controlled.
Therefore, an intelligent energy consumption scheduling device is needed, which not only can be used for adjusting and scheduling the use time point of electric equipment, the working time of staff and the number of people, but also can be used for monitoring and controlling the energy consumption of the device during valley so as to prevent the unexpected situation caused by too high energy consumption during valley.
Disclosure of Invention
The application aims to: aiming at the problems pointed out in the background art, the application provides an intelligent energy consumption scheduling method and device for a modern industrial production area, which are characterized in that an LSTM neural network is firstly used for acquiring data based on a data acquisition unit, then an AOA algorithm is used for carrying out energy consumption scheduling strategy design based on predicted data, the energy consumption of the industrial production area is scheduled, and on the premise of ensuring the normal operation of the production area, the economic benefit of the whole production area is improved, and the operation cost is reduced.
The technical scheme is as follows: the application provides an intelligent energy consumption adjusting method for a modern industrial production area, which comprises the following steps:
step 1: collecting equipment power consumption data, factory power load requirements, working time length data of staff in a production area, historical month asking data and staff number, wherein the equipment power consumption data comprise equipment power consumption data of an automatic production line, of staff participating in the production line, and carrying out normalization processing on the data;
step 2: based on factory area electricity load demand data and equipment electricity consumption data of an automatic production line, adopting an optimized LSTM neural network model to conduct automatic production line electricity consumption data prediction, and based on the electricity load demand data and the equipment electricity consumption data of the production line participated by staff, adopting the optimized LSTM neural network model to conduct production line electricity consumption data prediction participated by staff; the optimized LSTM neural network model specifically comprises the following steps: carrying out power consumption data prediction by adopting an LSTM neural network model, and simultaneously optimizing partial parameters of the LSTM neural network by adopting a badger optimization algorithm, wherein the partial parameters comprise the number of LSTM layers, the parameters of the Dense, the number of hidden layer neurons and the number of the neurons of the Dense layer;
step 3: and (3) according to the predicted automatic production line electricity consumption data in the step (2) and the equipment electricity consumption data of staff participating in the production line, solving a scheduling strategy conforming to the current situation by utilizing an AOA algorithm.
Further, the specific operation of performing the automatic line electricity data prediction by adopting the optimized LSTM neural network model in the step 2, and the line electricity data prediction participated by staff is as follows:
step 2.1: the preprocessed collected factory area electric load demand data and equipment electric power consumption data of an automatic production line or electric load demand data and equipment electric power consumption data of staff participating in the production line are used as input data;
step 2.2: dividing the data set into a training set and a testing set according to the ratio of 8:2;
step 2.3: the HBA algorithm is adopted to optimize parameters, and the specific steps are as follows:
step 2.3.1: initializing a population, and setting basic parameters of an algorithm;
x i ＝lb i +r 1 *(ub i -lb i )
where ub and lb are the search upper and lower boundaries, r 1 Is [0,1]Random number, x between i Is the ith individual of the population;
step 2.3.2: calculating the fitness value of each individual in the current population, and recording the current optimal fitness value and the corresponding individual position;
step 2.3.3: defining smell intensity, and updating density factors, wherein the smell intensity is related to concentration force of a prey and distance between badgers;
wherein I is i S is the source intensity, d i R is the distance between the prey and the current badger individual 2 Is [0,1]Random number, x between i And x i+1 Is the position of the individual meles, x prey Is a prey location;
wherein alpha is a density factor, C is a constant greater than or equal to 1, t is the current iteration number, t max The maximum iteration number;
step 2.3.4: updating the individual position, wherein the updating process is divided into two parts, namely an excavating stage and a honey collecting stage, judging through rand, and carrying out the excavating stage when rand is less than 0.5, and carrying out the honey collecting stage otherwise;
x new ＝x prey +F*β*I*x prey +F*r 3 *α*d i *|cos(2*π*r 4 )*[1-cos(2πr 5 )]|
x new ＝x prey +F*r 7 *α*d i
wherein x is new For the new position of the individual, x prey For the global optimal position, beta is more than or equal to 1 and represents the ability of the badger to acquire food, F is a sign for changing the searching direction, r 3 ,r 4 ,r 5 ,r 6 ,r 7 For different [0,1 ]]Random numbers in between;
step 2.3.5: calculating the fitness value again, and recording the optimal fitness value and the corresponding individual position;
step 2.3.6: judging whether the maximum iteration times are reached, if not, jumping to the step 2.3.3, and if so, outputting the optimal LSTM neural network parameters;
step 2.4: constructing an LSTM neural network model using optimized parameters, and performing model training by using training set data;
step 2.5: testing the test set data by the trained LSTM model, and calculating errors;
step 2.6: performing equipment power consumption data prediction by training;
step 2.7: and carrying out inverse normalization on the prediction result, and outputting predicted equipment power consumption data.
Further, the step 3 of solving the scheduling policy according with the current situation by using the AOA algorithm specifically includes the following steps:
step 3.1: initializing parameters and populations in an initializing algorithm;
step 3.2: calculating a fitness value, wherein a fitness function aims at the lowest cost;
wherein C is i The electricity cost of the ith electric equipment;
step 3.3: selecting a search stage according to a mathematical optimizer acceleration function, if r 1 >MOA, entering an exploration stage, jumping to the step 3.4, otherwise entering a development stage, and jumping to the step 3.5;
wherein r is 1 Is [0,1]The random numbers between Max and Min are the maximum value and the minimum value of the acceleration function respectively, T is the current iteration number, and T is the maximum iteration number;
step 3.4: the exploration stage realizes global search through multiplication operation and division operation, if r 2 >0.5, performing a division searching strategy, and otherwise performing a multiplication searching strategy;
wherein r is 2 Is [0,1]Random number in between, X (t+1) is the position of the candidate solution, X b (t) is the current optimal solution, MOP is the probability of a mathematical optimizer, epsilon is a minimum value, denominators are prevented from being 0, UB and LB are respectively the upper and lower bounds of the optimal solution, mu is a control parameter, and alpha is a sensitive parameter;
step 3.5: in the development stage, partial development is realized by utilizing addition operation and subtraction operation;
wherein r is 3 Is [0,1]Random numbers in between;
step 3.6, calculating the fitness value again, calculating a corresponding fitness value according to the updated individual position, and recording the optimal fitness value and the corresponding individual position;
step 3.7: judging whether the maximum iteration times are reached, if so, outputting the current optimal solution, further determining the running period of each device and the staff work system, and if not, returning to the step 3.3.
The application also discloses an intelligent energy consumption adjusting device for the modern industrial production area, which comprises a data acquisition unit, an energy consumption scheduling strategy design unit, an energy consumption scheduling unit and a scheduling object;
the data acquisition unit comprises an ammeter and staff work system collection unit; the energy consumption scheduling strategy design unit comprises a prediction module and a strategy design module; the dispatching object comprises electric equipment and staff in a factory;
the ammeter and employee work system collection unit in the data acquisition unit is connected with the prediction module in the energy consumption scheduling strategy design unit; the prediction module in the energy consumption scheduling strategy design unit is connected with the strategy design unit; the strategy design module in the energy consumption scheduling strategy design unit is connected with the energy consumption scheduling unit; the energy consumption scheduling unit is connected with electric equipment in the scheduling object and staff in a factory;
the data acquisition unit is used for acquiring equipment power utilization data, factory power load requirements, working time length data of staff in a production area, historical office data and staff number, wherein the equipment power utilization data comprises power utilization data of an automatic production line and power utilization data of staff participating in the production line;
the prediction module is used for predicting the power consumption data of the automatic production line by utilizing the optimized LSTM neural network model in the claim 1 based on the power load demand data and the power consumption data of the automatic production line equipment; based on the electricity load demand data and the electricity consumption data of the production line equipment participated by the staff, predicting the electricity consumption data of the production line participated by the staff by adopting the optimized LSTM neural network model in the claim 1; the strategy design module is used for receiving the prediction data and solving a scheduling strategy conforming to the current situation by utilizing the AOA algorithm in claim 1;
and the energy consumption scheduling unit performs energy consumption scheduling to the corresponding scheduling object based on the energy consumption scheduling strategy solved by the strategy design module.
Preferably, the energy consumption scheduling unit issues a power consumption scheduling instruction, schedules the power consumption in the scheduling object by controlling a power consumption using time point and a using time period, and simultaneously issues a scheduling instruction of the power consumption of the production line and the staff, and schedules staff in the production area by controlling the power consumption using time point, the using time period, the staff working time and the number of people.
The beneficial effects are that:
1. according to the electricity consumption data of each device, the electricity load demand of a factory and the electricity consumption data of each device, the data of an automatic production line and the data of staff participating in the production line are distinguished in sequence, the staff working system collecting unit records the working time length data, the historical month leave data and the staff number of staff in the production area, and the data are sent to the prediction module in the data acquisition unit through the data acquisition unit to conduct more specific and accurate prediction, so that the energy consumption can be monitored, and the use time point of electric equipment, the working time of staff and the number of staff can be regulated and controlled.
2. The energy consumption scheduling strategy design unit performs energy consumption scheduling based on the energy consumption scheduling strategy solved by the strategy design module, so that the production time length of the electric equipment is as long as possible in the valley time, the production time length is as short as possible in the peak time, less energy consumption is achieved, and accidents caused by excessive energy consumption can be avoided.
3. The application also adopts the LSTM neural network to predict the electricity consumption based on the data obtained by the data acquisition unit, comprises staff to participate in the electricity consumption prediction of the production line and the full-automatic production line, and adopts the AOA algorithm to design the energy consumption scheduling strategy based on the predicted data. And the energy consumption of the industrial production area is scheduled, so that the economic benefit of the whole production area is improved and the operation cost is reduced on the premise of ensuring the normal operation of the production area.
Drawings
FIG. 1 is a schematic diagram of a hardware architecture of the present application;
FIG. 2 is a flowchart of the HBA-LSTM algorithm employed in the present application;
FIG. 3 is a flow chart of an AOA algorithm employed by the present application;
FIG. 4 is a logic diagram of the operation of the present application;
FIG. 5 is a daily energy consumption comparison graph of the present application;
FIG. 6 is a graph comparing economic consumption of the present application over a year.
Detailed Description
The application is described in further detail below with reference to the accompanying drawings.
The application discloses an intelligent energy consumption adjusting device for a modern industrial production area, which is shown in fig. 1 and comprises a data acquisition unit 1, an energy consumption scheduling strategy design unit 2, an energy consumption scheduling unit 3 and a scheduling object 4. The data acquisition unit 1 comprises an ammeter and an employee work system collection unit. The energy consumption scheduling policy design unit 2 comprises a prediction module and a policy design module. The scheduling object 4 comprises electric equipment and staff in a factory.
The electricity meter and staff working system collecting unit in the data collecting unit 1 is connected with the prediction module in the energy consumption scheduling policy design unit, the policy prediction module in the energy consumption scheduling policy design unit 2 is connected with the policy design unit, the policy design module in the energy consumption scheduling policy design unit is connected with the energy consumption scheduling unit, and the energy consumption scheduling unit is connected with the electric equipment and staff in the scheduling object.
The ammeter records the electricity consumption data of each device and the electricity load demand of a factory, wherein the electricity consumption data of the device comprises the electricity consumption data of an automatic production line and the electricity consumption data of staff participating in the production line, and the staff working system collecting unit records the working time length data, the historical month leave data and the staff number of staff in the production area and sends the working time length data, the historical month leave data and the staff number to the prediction module through the data collecting unit 1.
The energy consumption scheduling strategy design unit 2 comprises a prediction module prediction and strategy design module, wherein the prediction module predicts the automatic production line electricity consumption data by adopting an optimized LSTM neural network model based on the electricity load demand data and the automatic production line equipment electricity consumption data, and predicts the production line electricity consumption data participated by staff by adopting the optimized LSTM neural network model based on the electricity load demand data and the production line equipment electricity consumption data participated by staff; and the strategy design module receives the prediction data and utilizes an AOA algorithm to solve a scheduling strategy conforming to the current situation.
The energy consumption scheduling strategy design unit 2 performs energy consumption scheduling based on the energy consumption scheduling strategy solved by the strategy design module, and aims to enable the production time of electric equipment to be as long as possible in valley time and to enable the production time to be as short as possible in peak time, so that less energy consumption and lower economic cost are achieved.
The scheduling policy designed by the policy design module in the energy consumption scheduling policy design unit 2 gives a scheduling instruction of the electric equipment, the electric equipment in the scheduling object is scheduled by controlling the using time point and the using time period of the electric equipment, meanwhile, staff participates in the scheduling instruction of the electric equipment and staff of the production line, and staff in the production area is scheduled by controlling the using time point and the using time period of the electric equipment, the working time of the staff and the number of people.
When the system is used, the electricity meter and the staff work system collecting unit in the data collecting unit 1 are connected with the prediction module, wherein the electricity meter records the electricity consumption data of each device and the electric load demand of a factory, the staff work system collecting unit records the working time length data, the historical month asking-for-the-false data and the staff number of staff in a production area, the electricity meter and the staff work system collecting unit send the electricity consumption data of the automatic production line device to the prediction module in the data collecting unit through the data collecting unit, the prediction module predicts the electricity consumption data of the automatic production line by adopting the optimized LSTM neural network model based on the electricity consumption data of the automatic production line device and the staff participating in the prediction module, the policy design module receives the prediction data and solves the scheduling policy conforming to the current situation by utilizing an AOA algorithm. And the energy consumption scheduling unit is connected with the electric equipment in the scheduling object and staff in the factory, controls the using time point of the electric equipment and the using time period to schedule the electric equipment in the scheduling object, and simultaneously, the staff participates in the scheduling of the electric equipment in the production line and the staff. Interference, such as "equipment related interference", "employee related interference" and "process related interference", may occur during operation of a general intelligent energy consumption device, but the present application can avoid related interference and perform energy consumption scheduling, so as to reduce energy consumption.
The optimized LSTM neural network model of the application carries out the prediction of the automatic production line electricity data and the production line electricity data participated by staff, and adopts a mel optimization algorithm to optimize partial parameters of the LSTM neural network, including LSTM layer number, dense parameter, hidden layer neuron number and Dense layer neuron number, so that the prediction model is more accurate and has lower error, and as shown in figure 2, the application mainly comprises the following steps:
1) Inputting data, namely inputting the collected factory electric load demand data and equipment electric data or electric load demand data of an automatic production line and equipment electric data of staff participating in the production line into a prediction model.
2) Data preprocessing, carrying out normalization processing on input data, wherein the formula is as follows:
wherein x is * To normalize the processed input data, x is the unprocessed input data, x min Is the minimum value, x, in the unprocessed input data max Is the maximum value in the unprocessed data.
3) The data set is divided into a training set and a testing set according to the ratio of 8:2.
4) The HBA algorithm is adopted to optimize parameters, and the specific steps are as follows:
4.1 Initializing population and setting basic parameters of algorithm;
x i ＝lb i +r 1 *(ub i -lb i )
where ub and lb are the search upper and lower boundaries, r 1 Is [0,1]Random number, x between i Is the ith individual of the population.
4.2 Calculating the fitness value of each individual in the current population, and recording the current optimal fitness value and the corresponding individual position.
4.3 The intensity of the smell is defined and updated, the intensity of the smell being related to the concentration of the prey and the distance between the badgers.
Wherein I is i S is the source intensity, d i R is the distance between the prey and the current badger individual 2 Is [0,1]Random number, x between i And x i+1 Is the position of the individual meles, x prey Is a hunting site.
Wherein alpha is a density factor, C is a constant greater than or equal to 1, t is the current iteration number, t max Is the maximum number of iterations.
4.4 Updating the individual position, wherein the updating process is divided into two parts, including an excavating stage and a honey collecting stage, the judgment is carried out through rand, when rand is less than 0.5, the excavating stage is carried out, and otherwise, the honey collecting stage is carried out;
x new ＝x prey +F*β*I*x prey +F*r 3 *α*d i *|cos(2*π*r 4 )*[1-cos(2πr 5 )]|
x new ＝x prey +F*r 7 *α*d i
wherein x is new For the new position of the individual, x prey For the global optimal position, beta is more than or equal to 1 and represents the ability of the badger to acquire food, F is a sign for changing the searching direction, r 3 ,r 4 ,r 5 ,r 6 ,r 7 For different [0,1 ]]Random numbers in between.
4.5 Again calculate fitness value and record the optimal fitness value, and the corresponding individual location.
4.6 If the maximum iteration number is not reached, jumping to the step 4.3, and outputting the optimal LSTM neural network parameters if the maximum iteration number is not reached.
5) And constructing an LSTM neural network model using the optimized parameters, and performing model training by using training set data.
6) And testing the test set data by the trained LSTM model, and calculating errors.
7) The training is used to make equipment power consumption data predictions.
8) And carrying out inverse normalization on the prediction result, and outputting predicted equipment power consumption data.
The population of the prediction method in the embodiment of the application is the number of LSTM layers, the response parameters, the number of hidden layer neurons and the number of response layer neurons which can achieve the purpose of prediction, and the value of the fitness refers to the prediction accuracy; the purpose of the algorithm is to find a solution for a set of neural network parameters that maximizes the accuracy of the prediction.
The electric equipment is operated and scheduled by adopting an AOA optimization algorithm, and a reasonable staff work system is designed, as shown in fig. 3, the implementation process is specifically as follows:
1) Initializing parameters and populations in an initialization algorithm.
2) Calculating a fitness value, wherein a fitness function aims at the lowest cost:
wherein C is i The electricity cost of the ith electric equipment;
3) Selecting a search stage according to a mathematical optimizer acceleration function, if r 1 >MOA, entering an exploration stage, jumping to the step 4), otherwise entering a development stage, jumping to the step 5).
Wherein r is 1 Is [0,1]The random numbers, max and Min, are the maximum value and the minimum value of the acceleration function respectively, T is the current iteration number, and T is the maximum iteration number.
4) Search stepSegment, the exploration stage realizes global search through multiplication operation and division operation, if r 2 >0.5, performing a division search strategy, and otherwise performing a multiplication search strategy:
wherein r is 2 Is [0,1]Random number in between, X (t+1) is the position of the candidate solution, X b (t) is the current optimal solution, MOP is the probability of a mathematical optimizer, epsilon is the minimum value, denominator is 0, UB and LB are the upper and lower bounds of the optimal solution respectively, mu is a control parameter, and alpha is a sensitive parameter.
5) In the development stage, partial development is realized by utilizing addition operation and subtraction operation:
wherein r is 3 Is [0,1]Random numbers in between.
6) And calculating the fitness value again, calculating a corresponding fitness value according to the updated individual position, and recording the optimal fitness value and the corresponding individual position.
7) Step 3.7: judging whether the maximum iteration times are reached, if so, outputting the current optimal solution, further determining the running period of each device and the staff work system, and if not, returning to the step 3).
Aiming at the operation scheduling optimization method, the individual with the best fitness value found in the embodiment of the application is the individual with the best cost, and the fitness value mentioned in the algorithm is the cost; the population of the algorithm refers to preprocessed collected factory electricity load demand data, equipment electricity consumption data or electricity load demand data of an automatic production line and equipment electricity consumption data of staff participating in the production line, and the lowest cost meeting the demand is found through the data.
As shown in fig. 5, in the daily energy consumption comparison of the application and other regulating equipment, the daily energy consumption of a factory is lower than that of the factory after the use of the other regulating equipment by adopting the regulating device, the regulating device is focused on the energy consumption in a non-processing process, and unnecessary energy consumption is reduced as much as possible, and compared with the regulating device, the effect of the regulating device is excellent.
As shown in fig. 6, in comparison with the annual economic cost of other regulating devices, the regulating device of the application reduces the economic cost of a factory, and reduces the economic cost by ensuring that the working time is as long as possible under the condition of low energy consumption.
The foregoing embodiments are merely illustrative of the technical concept and features of the present application, and are intended to enable those skilled in the art to understand the present application and to implement the same, not to limit the scope of the present application. All equivalent changes or modifications made according to the spirit of the present application should be included in the scope of the present application.
Claims (5)
1. The intelligent energy consumption regulating method for the modern industrial production area is characterized by comprising the following steps of:
step 1: collecting equipment power consumption data, factory power load requirements, working time length data of staff in a production area, historical month asking data and staff number, wherein the equipment power consumption data comprise equipment power consumption data of an automatic production line, of staff participating in the production line, and carrying out normalization processing on the data;
step 2: based on factory area electricity load demand data and equipment electricity consumption data of an automatic production line, adopting an optimized LSTM neural network model to conduct automatic production line electricity consumption data prediction, and based on the electricity load demand data and the equipment electricity consumption data of the production line participated by staff, adopting the optimized LSTM neural network model to conduct production line electricity consumption data prediction participated by staff; the optimized LSTM neural network model specifically comprises the following steps: carrying out power consumption data prediction by adopting an LSTM neural network model, and simultaneously optimizing partial parameters of the LSTM neural network by adopting a badger optimization algorithm, wherein the partial parameters comprise the number of LSTM layers, the parameters of the Dense, the number of hidden layer neurons and the number of the neurons of the Dense layer;
step 3: and (3) according to the predicted automatic production line electricity consumption data in the step (2) and the equipment electricity consumption data of staff participating in the production line, solving a scheduling strategy conforming to the current situation by utilizing an AOA algorithm.
2. The intelligent energy consumption adjusting method for the modern industrial production area according to claim 1, wherein the specific operations of the automatic production line electricity consumption data and the production line electricity consumption data prediction participated by staff by adopting the optimized LSTM neural network model in the step 2 are as follows:
step 2.1: the preprocessed collected factory area electric load demand data and equipment electric power consumption data of an automatic production line or electric load demand data and equipment electric power consumption data of staff participating in the production line are used as input data;
step 2.2: dividing the data set into a training set and a testing set according to the ratio of 8:2;
step 2.3: the HBA algorithm is adopted to optimize parameters, and the specific steps are as follows:
step 2.3.1: initializing a population, and setting basic parameters of an algorithm:
x i ＝lb i +r 1 *(ub i -lb i )
where ub and lb are the search upper and lower boundaries, r 1 Is [0,1]Random number, x between i Is the ith individual of the population; the population is the number of LSTM layers, the response parameters, the number of hidden layer neurons and the number of response layer neurons of each group which can achieve the prediction purpose;
step 2.3.2: calculating the fitness value of each individual in the current population, and recording the current optimal fitness value and the corresponding individual position, wherein the fitness value is the prediction accuracy;
step 2.3.3: the intensity of the smell is defined and updated, and the intensity of the smell is related to the concentration of the prey and the distance between the badgers:
wherein I is i S is the source intensity, d i R is the distance between the prey and the current badger individual 2 Is [0,1]Random number, x between i And x i+1 Is the position of the individual meles, x prey Is a prey location;
wherein alpha is a density factor, C is a constant greater than or equal to 1, t is the current iteration number, t max The maximum iteration number;
step 2.3.4: updating the individual position, wherein the updating process is divided into two parts, namely an excavating stage and a honey collecting stage, judging through rand, and carrying out the excavating stage when rand is less than 0.5, and carrying out the honey collecting stage otherwise;
x new ＝x prey +F*β*I*x prey +F*r 3 *α*d i *|cos(2*π*r 4 )*[1-cos(2πr 5 )]|
x new ＝x prey +F*r 7 *α*d i
wherein x is new For the new position of the individual, x prey For the global optimal position, beta is more than or equal to 1 and represents the ability of the badger to acquire food, F is a sign for changing the searching direction, r 3 ,r 4 ,r 5 ,r 6 ,r 7 For different [0,1 ]]Random numbers in between;
step 2.3.5: calculating the fitness value again, and recording the optimal fitness value and the corresponding individual position;
step 2.3.6: judging whether the maximum iteration times are reached, if not, jumping to the step 2.3.3, and if so, outputting the optimal LSTM neural network parameters;
step 2.4: constructing an LSTM neural network model using optimized parameters, and performing model training by using training set data;
step 2.5: testing the test set data by the trained LSTM model, and calculating errors;
step 2.6: performing equipment power consumption data prediction by training;
step 2.7: and carrying out inverse normalization on the prediction result, and outputting predicted equipment power consumption data.
3. The intelligent energy consumption adjusting method for modern industrial production areas according to claim 1, wherein the step 3 of solving the scheduling policy according to the current situation by using the AOA algorithm specifically comprises the following steps:
step 3.1: initializing parameters and populations in an algorithm, wherein the populations refer to predicted automatic production line electricity data and equipment electricity data of staff participating in the production line, and searching the lowest cost meeting the requirements through the data;
step 3.2: calculating a fitness value, wherein a fitness function aims at the lowest cost;
wherein C is i The electricity cost of the ith electric equipment;
step 3.3: selecting a search stage according to a mathematical optimizer acceleration function, if r 1 >MOA, entering an exploration stage, jumping to the step 3.4, otherwise entering a development stage, and jumping to the step 3.5;
wherein r is 1 Is [0,1]The random numbers between Max and Min are the maximum value and the minimum value of the acceleration function respectively, T is the current iteration number, and T is the maximum iteration number;
step 3.4: the exploration stage realizes global search through multiplication operation and division operation, if r 2 >0.5, performing a division searching strategy, and otherwise performing a multiplication searching strategy;
wherein r is 2 Is [0,1]Random number in between, X (t+1) is the position of the candidate solution, X b (t) is the current optimal solution, MOP is the probability of a mathematical optimizer, epsilon is a minimum value, denominators are prevented from being 0, UB and LB are respectively the upper and lower bounds of the optimal solution, mu is a control parameter, and alpha is a sensitive parameter;
step 3.5: in the development stage, partial development is realized by utilizing addition operation and subtraction operation;
wherein r is 3 Is [0,1]Random numbers in between;
step 3.6, calculating the fitness value again, calculating the corresponding fitness value according to the updated individual position, and recording the optimal fitness value and the corresponding individual position, wherein the individual with the optimal fitness value is the individual with the optimal cost;
step 3.7: judging whether the maximum iteration times are reached, outputting the current optimal solution if the maximum iteration times are reached, determining the running period of each device and the staff work system, and returning to the step 3.3 if the maximum iteration times are not reached.
4. A modern industrial production area intelligent energy consumption regulating device based on the modern industrial production area intelligent energy consumption regulating method as claimed in claim 1, which is characterized by comprising a data acquisition unit (1), an energy consumption scheduling strategy design unit (2), an energy consumption scheduling unit (3) and a scheduling object (4);
the data acquisition unit (1) comprises an ammeter and staff work system collection unit; the energy consumption scheduling strategy design unit (2) comprises a prediction module and a strategy design module; the scheduling object (4) comprises electric equipment and factory staff;
the ammeter and employee work system collection unit in the data acquisition unit (1) is connected with the prediction module in the energy consumption scheduling strategy design unit (2); the prediction module in the energy consumption scheduling strategy design unit (2) is connected with the strategy design unit; the energy consumption scheduling strategy design module in the energy consumption scheduling strategy design unit (2) is connected with the energy consumption scheduling unit (3); the energy consumption scheduling unit (3) is connected with electric equipment and staff in a factory area in the scheduling object (4);
the data acquisition unit (1) is used for acquiring equipment power consumption data, factory power load requirements, working time length data of staff in a production area, historical month leave data and staff number, wherein the equipment power consumption data comprise power consumption data of an automatic production line and power consumption data of staff participating in the production line;
the prediction module is used for predicting the power consumption data of the automatic production line by utilizing the optimized LSTM neural network model based on the power load demand data and the power consumption data of the automatic production line equipment; based on the electricity load demand data and the electricity consumption data of the production line equipment participated by the staff, predicting the electricity consumption data of the production line participated by the staff by utilizing the optimized LSTM neural network model; the strategy design module receives the prediction data and utilizes an AOA algorithm to solve a scheduling strategy conforming to the current situation;
and the energy consumption scheduling unit (3) performs energy consumption scheduling to the corresponding scheduling object (4) based on the energy consumption scheduling strategy solved by the strategy design module.
5. The intelligent energy consumption adjusting device for the modern industrial production area according to claim 4, wherein the energy consumption scheduling unit (3) issues an electric equipment scheduling instruction, schedules the electric equipment in the scheduling object (4) by controlling the use time point and the use time period of the electric equipment, issues the electric equipment and the staff scheduling instruction of the staff participating in the production line, and schedules the staff in the production area by controlling the use time point and the use time period of the electric equipment, the staff working time and the number of people.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310236536.2A CN116070881B (en) | 2023-03-13 | 2023-03-13 | Intelligent energy consumption scheduling method and device for modern industrial production area |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310236536.2A CN116070881B (en) | 2023-03-13 | 2023-03-13 | Intelligent energy consumption scheduling method and device for modern industrial production area |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116070881A CN116070881A (en) | 2023-05-05 |
| CN116070881B true CN116070881B (en) | 2023-09-29 |
Family
ID=86178737
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310236536.2A Active CN116070881B (en) | 2023-03-13 | 2023-03-13 | Intelligent energy consumption scheduling method and device for modern industrial production area |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116070881B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116562444A (en) * | 2023-05-11 | 2023-08-08 | 淮阴工学院 | Intelligent carbon emission adjusting method and equipment for industrial park |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109784532A (en) * | 2018-12-12 | 2019-05-21 | 中山大学 | A kind of plant area's energy consumption prediction technique and system based on deep learning |
| CN112561728A (en) * | 2020-10-28 | 2021-03-26 | 西安交通大学 | Attention mechanism LSTM-based comprehensive energy consumption cost optimization method, medium and equipment |
| CN114742278A (en) * | 2022-03-17 | 2022-07-12 | 西安建筑科技大学 | Building energy consumption prediction method and system based on improved LSTM |
| CN115186864A (en) * | 2022-05-23 | 2022-10-14 | 国网浙江省电力有限公司经济技术研究院 | System and method for managing energy efficiency of small micro-park based on digital twin technology |
| WO2023010624A1 (en) * | 2021-08-04 | 2023-02-09 | 浙江大学 | Wind turbine layout optimization method integrating wind farm scheduling policy |
| CN115758912A (en) * | 2022-12-08 | 2023-03-07 | 上海电机学院 | Air conditioner energy consumption optimizing system |
-
2023
- 2023-03-13 CN CN202310236536.2A patent/CN116070881B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109784532A (en) * | 2018-12-12 | 2019-05-21 | 中山大学 | A kind of plant area's energy consumption prediction technique and system based on deep learning |
| CN112561728A (en) * | 2020-10-28 | 2021-03-26 | 西安交通大学 | Attention mechanism LSTM-based comprehensive energy consumption cost optimization method, medium and equipment |
| WO2023010624A1 (en) * | 2021-08-04 | 2023-02-09 | 浙江大学 | Wind turbine layout optimization method integrating wind farm scheduling policy |
| CN114742278A (en) * | 2022-03-17 | 2022-07-12 | 西安建筑科技大学 | Building energy consumption prediction method and system based on improved LSTM |
| CN115186864A (en) * | 2022-05-23 | 2022-10-14 | 国网浙江省电力有限公司经济技术研究院 | System and method for managing energy efficiency of small micro-park based on digital twin technology |
| CN115758912A (en) * | 2022-12-08 | 2023-03-07 | 上海电机学院 | Air conditioner energy consumption optimizing system |
Non-Patent Citations (1)
| Title |
|---|
| 基于ARIMA LSTM组合模型的楼宇短期负荷预测方法研究;李鹏辉;崔承刚;杨宁;陈辉;;上海电力学院学报(第06期);67-73 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116070881A (en) | 2023-05-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116646933B (en) | Big data-based power load scheduling method and system | |
| CN113962364B (en) | Multi-factor power load prediction method based on deep learning | |
| CN110659779B (en) | Power distribution system network loss prediction method based on long-term and short-term memory network | |
| CN111353656B (en) | Steel enterprise oxygen load prediction method based on production plan | |
| CN103295075B (en) | A kind of ultra-short term load forecast and method for early warning | |
| CN116070881B (en) | Intelligent energy consumption scheduling method and device for modern industrial production area | |
| CN104009494A (en) | Environmental economy power generation dispatching method | |
| CN109647899A (en) | More specification rolled piece power consumption forecasting procedures in a kind of hot strip rolling finishing stands | |
| CN113902582A (en) | Building comprehensive energy load prediction method and system | |
| CN114519451B (en) | Intelligent island type park vehicle carbon emission prediction method and system | |
| CN116581750A (en) | Intelligent line load charging method based on power grid load level | |
| Yahya et al. | Short-term electric load forecasting using recurrent neural network (study case of load forecasting in central java and special region of yogyakarta) | |
| CN111313449B (en) | Cluster electric vehicle power optimization management method based on machine learning | |
| CN110866633A (en) | Micro-grid ultra-short term load prediction method based on SVR support vector regression | |
| CN116109095A (en) | Day-ahead optimal scheduling method and system for heating system considering supply and demand coordination | |
| CN113568311B (en) | Knowledge information-based intelligent optimal control method for sewage treatment | |
| CN112950001A (en) | Intelligent energy management and control system and method based on cloud edge closed-loop architecture | |
| CN117113086A (en) | Energy storage unit load prediction method, system, electronic equipment and medium | |
| KR102556093B1 (en) | Reward generation method to reduce peak load of electric power and action control apparatus performing the same method | |
| CN114372691A (en) | Electric energy substitution potential estimation method based on holographic perception | |
| CN113283638A (en) | Load extreme curve prediction method and system based on fusion model | |
| CN112561115A (en) | Active power distribution network user power consumption behavior variable weight combination prediction model and prediction method | |
| CN117726034A (en) | Gas parallel prediction method and system based on double models and electronic equipment | |
| Xie et al. | A short-term power load prediction model based on improved wavelet neural networks | |
| CN117374963A (en) | Ultra-short-term power demand prediction method for iron and steel enterprises based on load characteristics |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |