CN117751373A - Method and device for predicting factory operation and computer readable storage medium - Google Patents
Method and device for predicting factory operation and computer readable storage medium Download PDFInfo
- Publication number
- CN117751373A CN117751373A CN202180101142.5A CN202180101142A CN117751373A CN 117751373 A CN117751373 A CN 117751373A CN 202180101142 A CN202180101142 A CN 202180101142A CN 117751373 A CN117751373 A CN 117751373A
- Authority
- CN
- China
- Prior art keywords
- production process
- energy consumption
- consumption data
- electric energy
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 67
- 238000003860 storage Methods 0.000 title claims abstract description 17
- 238000004519 manufacturing process Methods 0.000 claims abstract description 244
- 238000005265 energy consumption Methods 0.000 claims abstract description 155
- 230000008569 process Effects 0.000 claims description 17
- 239000010865 sewage Substances 0.000 claims description 16
- 238000003062 neural network model Methods 0.000 claims description 13
- 238000000926 separation method Methods 0.000 claims description 13
- 230000015654 memory Effects 0.000 claims description 11
- 238000009826 distribution Methods 0.000 claims description 10
- 238000012549 training Methods 0.000 claims description 9
- 238000000354 decomposition reaction Methods 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 6
- 238000010586 diagram Methods 0.000 description 11
- 238000013528 artificial neural network Methods 0.000 description 7
- 230000005611 electricity Effects 0.000 description 5
- 238000011156 evaluation Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 239000002351 wastewater Substances 0.000 description 5
- 238000010276 construction Methods 0.000 description 4
- 238000005259 measurement Methods 0.000 description 4
- 238000004378 air conditioning Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000003556 assay Methods 0.000 description 2
- 125000004122 cyclic group Chemical group 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 230000007812 deficiency Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000008713 feedback mechanism Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012905 input function Methods 0.000 description 1
- 239000013067 intermediate product Substances 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000006386 memory function Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000003345 natural gas Substances 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000005309 stochastic process Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Computational Linguistics (AREA)
- Development Economics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Game Theory and Decision Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention discloses a prediction method and device for factory operation and a computer readable storage medium. The method comprises the following steps: separating the electric energy consumption data of the workshop in the production process from the electric energy consumption data of the workshop in the factory; determining an electric energy consumption prediction model adapted to predict electric energy consumption data in the production process based on the electric energy consumption data in the production process; and predicting the electric energy consumption data of the workshop in a future production process based on the electric energy consumption prediction model. For factories which do not provide data such as production conditions and production predictions, the operation conditions of the factories can be predicted based on the electric energy consumption data in the production process, so that various services such as energy services or logistics services can be conveniently provided.
Description
The invention relates to the technical field of factory data processing, in particular to a method and a device for predicting factory operation and a computer readable storage medium.
The construction of intelligent industrial parks is an important research field which is of great concern in the operation and management of modern industrial parks, and the technology is expanding and maturing increasingly. With the development of technology, the demands for developing intelligent industrial parks are increasing, so as to provide better services for the enterprises of the parks. The intelligent park construction comprises energy supply, traffic logistics, security, environmental protection, industry coordination and the like. The intelligence of these panels is often supported by the current state of production and production planning data of each enterprise. However, in real world situations, enterprises often do not provide data such as production conditions and production predictions for business privacy or other purposes.
At present, aiming at the problems of production state judgment, energy service, logistics service, metering station use plan and the like in a large scene of the deficiency of production data of a campus enterprise, an accurate method and model with pertinence and high efficiency are still lacking, and the requirement of intelligent campus construction on acquiring accurate production data of the enterprise is difficult to meet.
Disclosure of Invention
The embodiment of the invention provides a prediction method and device for factory operation and a computer readable storage medium.
A method of predicting plant operation, the method comprising:
separating the electric energy consumption data of the workshop in the production process from the electric energy consumption data of the workshop in the factory;
determining an electric energy consumption prediction model adapted to predict electric energy consumption data in the production process based on the electric energy consumption data in the production process;
and predicting the electric energy consumption data of the workshop in a future production process based on the electric energy consumption prediction model.
It can be seen that for factories lacking enterprise production data, the electrical energy consumption data in future production processes can be predicted based on the electrical energy consumption data in the production processes, thereby facilitating the provision of energy services.
In one embodiment, the separating the power consumption data of the plant during the production process from the power consumption data of the plant in the plant includes:
separating the electric energy consumption data of the workshop in the production process from the electric energy consumption data of the workshop in the factory based on an energy decomposition mode of a non-parametric factor hidden Markov model; or (b)
The power consumption data of the plant during production is separated from the power consumption data of the plant in the plant based on the percentages determined by the historical power distribution data.
Therefore, by separating out the electrical energy consumption data in the production process, the plant operating conditions can be predicted more accurately.
In one embodiment, the method further comprises:
determining an operating condition of the production process based on the electrical energy consumption data in the production process, wherein:
when the electric energy consumption data is smaller than a preset first threshold value, determining that the working condition of the production process is a stop condition;
when the electric energy consumption data is larger than the first threshold value and smaller than a preset second threshold value, determining that the working condition of the production process is a waiting condition;
when the electric energy consumption data is larger than the second threshold value, determining that the working condition of the production process is an active running condition;
wherein the second threshold value is greater than the first threshold value.
It can be seen that the operating conditions of the production process can be determined based on the power consumption data of the production process, thereby facilitating understanding of the operating conditions.
In one embodiment, the method further comprises: determining a yield of the production process and a logistics demand of the production process based on the electrical energy consumption data in the production process;
wherein the method comprises the steps of
Wherein the yield of the production process is Qk (t); the logistic requirement of the production process is Cargo k (t); cap is the transport capacity of each transport; p (P) k (t) is electrical energy consumption data in the production process; q (Q) krated Rated capacity for the production process; p (P) krated Is the rated electrical power of the generation process; t is a time parameter.
Thus, the output and logistics demand can be determined based on the power consumption data of the production process, facilitating the provision of logistics services.
In one embodiment, the determining the power consumption prediction model adapted to predict the power consumption data in the production process based on the power consumption data in the production process comprises:
the electric energy consumption data in the production process, the steam consumption in the production process and the model input data are used as training data together to train the neural network model so as to obtain the electric energy consumption prediction model; wherein the model input data comprises at least one of: a time stamp; weather; a temperature; sewage output; a sewage test result; working conditions; and (5) a holiday.
Thus, a power consumption prediction model may be established based on artificial intelligence techniques. In particular, the trained neural network model is more accurate using the determined operating conditions as model input data.
In one embodiment, the predicting the electrical energy consumption data of the plant in a future production process based on the electrical energy consumption prediction model includes:
inputting model input data in a future production process into the electric energy consumption prediction model to predict electric energy consumption data in the future production process and steam consumption data in the future production process; the method further comprises the steps of:
determining at least one of the following based on the electrical energy consumption data in the future production process: predicted production of future production processes; predicted logistics requirements for future production processes; predicted storage requirements for future production processes.
Therefore, the electric energy consumption data in the future production process is predicted based on the electric energy consumption prediction model, so that logistics service and storage service which need to be provided in the future can be conveniently planned. Moreover, steam service to be provided in the future is conveniently planned based on the steam consumption data in the future production process.
A plant operation prediction apparatus, comprising:
the separation module is used for separating the electric energy consumption data of the workshop in the production process from the electric energy consumption data of the workshop in the factory;
a determining module for determining an electric energy consumption prediction model adapted to predict electric energy consumption data in the production process based on the electric energy consumption data in the production process;
and the prediction module is used for predicting the electric energy consumption data of the workshop in the future production process based on the electric energy consumption prediction model.
It can be seen that for factories lacking enterprise production data, the electrical energy consumption data in future production processes can be predicted based on the electrical energy consumption data in the production processes, thereby facilitating the provision of energy services.
In one embodiment, the separation module is configured to separate the electrical energy consumption data of the plant in the production process from the electrical energy consumption data of the plant in the factory based on an energy decomposition mode of a non-parametric factor hidden markov model; or, based on the percentage determined by the historical power distribution data, separating the power consumption data of the plant during the production process from the power consumption data of the plant in the plant.
Therefore, by separating out the electrical energy consumption data in the production process, the plant operating conditions can be predicted more accurately.
In one embodiment, the determining module is further configured to determine an operating condition of the production process based on the electrical energy consumption data in the production process, wherein: when the electric energy consumption data is smaller than a preset first threshold value, determining that the working condition of the production process is a stop condition; when the electric energy consumption data is larger than the first threshold value and smaller than a preset second threshold value, determining that the working condition of the production process is a waiting condition; when the electric energy consumption data is larger than the second threshold value, determining that the working condition of the production process is an active running condition; wherein the second threshold value is greater than the first threshold value.
It can be seen that the operating conditions of the production process can be determined based on the power consumption data of the production process, thereby facilitating understanding of the operating conditions.
In one embodiment, the determining module is further configured to determine a yield of the production process and a logistic demand of the production process based on the electrical energy consumption data in the production process;
wherein the method comprises the steps of
Wherein the yield of the production process is Q k (t); the logistic requirement of the production process is Cargo k (t); cap is the transport capacity of each transport; p (P) k (t) is electrical energy consumption data in the production process; q (Q) krated Rated capacity for the production process; p (P) krated Is the rated electrical power of the generation process; t is a time parameter.
Thus, the output and logistics demand can be determined based on the power consumption data of the production process, facilitating the provision of logistics services.
In one embodiment, the determining module is configured to use the electric energy consumption data in the production process, the steam consumption in the production process, and the model input data together as training data to train the neural network model to obtain the electric energy consumption prediction model; wherein the model input data comprises at least one of: a time stamp; weather; a temperature; sewage output; a sewage test result; and (5) a holiday.
Thus, a power consumption prediction model may be established based on artificial intelligence techniques. In particular, the trained neural network model is more accurate using the determined operating conditions as model input data.
In one embodiment, the prediction module is configured to input model input data in a future production process to the electric energy consumption prediction model to predict electric energy consumption data in the future production process and steam consumption data in the future production process; determining at least one of the following based on the electrical energy consumption data in the future production process: predicted production of future production processes; predicted logistics requirements for future production processes; predicted storage requirements for future production processes.
Therefore, the electric energy consumption data in the future production process is predicted based on the electric energy consumption prediction model, so that logistics service and storage service which need to be provided in the future can be conveniently planned. Moreover, steam service to be provided in the future is conveniently planned based on the steam consumption data in the future production process.
A prediction apparatus for plant operation, comprising a processor and a memory;
the memory has stored therein an application executable by the processor for causing the processor to perform the method of predicting plant operation as described in any one of the above.
A computer readable storage medium having stored therein computer readable instructions for performing the method of predicting plant operation as described in any one of the above.
The above and other features and advantages of the present invention will become more apparent to those of ordinary skill in the art by describing in detail preferred embodiments thereof with reference to the attached drawings in which:
fig. 1 is a flowchart of a method of predicting plant operation according to an embodiment of the present invention.
Fig. 2 is an exemplary schematic diagram of a method of predicting plant operation in accordance with an embodiment of the present invention.
FIG. 3 is an exemplary schematic diagram of a non-parametric, factor hidden Markov model (NFHMM) process of an embodiment of the invention.
Fig. 4 is an exemplary structural diagram of a Recurrent Neural Network (RNN) according to an embodiment of the present invention.
Fig. 5 is a block diagram of a plant operation prediction apparatus according to an embodiment of the present invention.
Fig. 6 is a block diagram of a plant-operated prediction apparatus having a processor-memory architecture according to an embodiment of the present invention.
Wherein, the reference numerals are as follows:
| reference numerals | Meaning of |
| 100 | Prediction method for factory operation |
| 101～103 | Step (a) |
| 20 | Identification process of production process |
| 211、221、2m1 | Electric energy consumption data of workshops |
| 212、222、2m2 | Separation process |
| 213、223、2m3 | Production assessment |
| 40 | Electric energy consumption prediction model |
| 41 | Service |
| 31 | Steam consumption, sewage yield, sewage assay results, etc |
| 32 | Factory provided production process statistics |
| 50 | Prior distribution |
| 51 | Super parameter |
| 52 | Conversion matrix parameters |
| 53 | Device status |
| 54 | Device energy consumption characteristics |
| 55 | Electric energy measurement |
| 500 | Prediction device for factory operation |
| 501 | Separation module |
| 502 | Determination module |
| 503 | Prediction module |
| 600 | Prediction device for factory operation |
| 601 | Processor and method for controlling the same |
| 602 | Memory device |
The present invention will be further described in detail with reference to the following examples, in order to make the objects, technical solutions and advantages of the present invention more apparent.
For simplicity and clarity of description, the following description sets forth aspects of the invention by describing several exemplary embodiments. Numerous details in the embodiments are provided solely to aid in the understanding of the invention. It will be apparent, however, that the embodiments of the invention may be practiced without limitation to these specific details. Some embodiments are not described in detail in order to avoid unnecessarily obscuring aspects of the present invention, but rather only to present a framework. Hereinafter, "comprising" means "including but not limited to", "according to … …" means "according to at least … …, but not limited to only … …". The term "a" or "an" is used herein to refer to a number of components, either one or more, or at least one, unless otherwise specified.
In consideration of the problems of production state judgment, energy service, logistics service, metering station use plan and the like in the large scene of the deficiency of the production state of a park enterprise at present, the method and model which are specific and efficient are still lacking, and the current situation that the intelligent park construction is difficult to meet the requirement of acquiring accurate enterprise production data is difficult to meet.
Fig. 1 is a flowchart of a method of predicting plant operation according to an embodiment of the present invention.
As shown in fig. 1, the method 100 includes:
step 101: and separating the electric energy consumption data of the workshop in the production process from the electric energy consumption data of the workshop in the factory.
In practice, the campus is often only able to measure the power consumption of each production plant (or called section) without detailed power consumption data for each device in each enterprise plant. Meanwhile, depending on the power consumption of each production plant, there may be a case where two or more plants share one measurement point.
First, based on the expert knowledge of the plant production characteristics, a production plant, i.e., the "plant" described in step 101, is judged to be functionally critical and can efficiently collect energy consumption data. The electricity consumption data of the plant mentioned in step 101 also typically includes portions that are not directly produced by the reaction, such as lighting, air conditioning, maintenance, etc., and thus it is necessary to separate the electricity consumption data of the portions that are not directly produced by the reaction in step 101.
In one embodiment, the power consumption data of the workshop during the production process is separated from the power consumption data of the workshop in the factory based on the power decomposition method of a non-parametric factor hidden Markov model (NFHMM). A hidden markov model is a type of markov chain whose states are not directly observable, but are observable through a sequence of observation vectors, each of which is represented as various states by some probability density distribution, each of which is generated by a sequence of states having a corresponding probability density distribution. Therefore, the hidden Markov model is a double stochastic process, with hidden Markov chains of a certain state number and a set of display stochastic functions.
The energy decomposition mode of NFHMM is described in detail below. Fig. 3 is an exemplary schematic diagram of NFHMM processing in accordance with an embodiment of the present invention.
Suppose that an aggregate energy consumption signal Y (T) is obtained for K devices (or processes) of a plant-wide plant, where T is from 1 to T. The K devices comprise devices for generating processes, and devices for lighting, air conditioning, overhaul and the like which do not directly react for production. Here, no a priori knowledge 50 about the number of devices is assumed, i.e. K is unknown. Let Z be T.times.K matrix, and Z corresponds to (T, K) term t，k Indicating the state of device k at time t. In fig. 3, the NFHMM is described as follows:
assume that the state of each device evolves according to the transition matrix parameters 52:
the method comprises the steps of carrying out a first treatment on the surface of the Wherein the method comprises the steps ofState z of kth device at time t t，k FollowPut the a priori distribution 50 at μ k ～Beta(α/K，1)，b k Beta (gamma, delta). The super-parameters 51 contain α, γ, δ, and H. The device state 53 contains the state of each device. The device energy consumption feature 54 is Θ, where the kth entry of the vector represents the kthThe gaussian distributed power of the device. The observed electrical energy measurement 55 (mixed signal Y) is generated by the emission model y=zΘ+epsilon, where epsilon is the measurement noise.
While the above exemplary descriptions of typical examples of using NFHMM to process power consumption data in a separation plant during production, those skilled in the art will recognize that this description is exemplary only and is not intended to limit the scope of embodiments of the present invention.
In one embodiment, the power consumption data of the plant during production is separated from the power consumption data of the plant within the plant based on a percentage determined from the historical power distribution data.
In one embodiment, the operating condition of the production process is determined based on the electrical energy consumption data in the production process, wherein: when the electric energy consumption data is smaller than a preset first threshold value, determining that the working condition of the production process is a stop condition; when the electric energy consumption data is larger than the first threshold value and smaller than a preset second threshold value, determining that the working condition of the production process is a waiting condition; when the electric energy consumption data is larger than the second threshold value, determining that the working condition of the production process is an active running condition; wherein the second threshold value is greater than the first threshold value.
Specifically, C k (t)＝Condition_map(P k (t))；
C k And (t) is an operating condition. An exemplary matching method is as follows, and other matching methods may be employed: if P k ＜P 1 The workshop is in production stopping; if P 1 ＜P k ＜P 2 The workshop is in production waiting; if P 2 ＜P k The workshop is in process.
Moreover, according to the production working condition of each workshop, logistics and the like can be reasonably planned. Moreover, based on the decomposed power consumption, the output of the enterprise can be calculated according to the relation between the industry power consumption and the output. Based on the obtained power consumption of the device and the possible working conditions of the device, the power consumption can be matched with the working condition mode of the device. The production rule of the enterprise can be obtained based on the working conditions and duration of each link. The relation among the key processes is mined, and the meaning of the relation to energy consumption is as follows: the consumption data and the duration of each flow are converted into the output of various intermediate products, the stock of a warehouse, the corresponding logistics requirements and the like.
In one embodiment, the method further comprises: determining a yield of the production process and a logistics demand of the production process based on the electrical energy consumption data in the production process;
wherein the method comprises the steps of
Wherein the yield of the production process is Q k (t); the logistic requirement of the production process is Cargo k (t); cap is the transport capacity of each transport; p (P) k (t) is electrical energy consumption data in the production process; q (Q) krated Rated capacity for the production process; t is a time parameter.
Step 102: an electrical energy consumption prediction model adapted to predict electrical energy consumption data during production is determined based on the electrical energy consumption data.
The electric energy consumption data in the production process, the steam consumption in the production process and the model input data are used as training data together to train the neural network model so as to obtain the electric energy consumption prediction model; wherein the model input data includes at least one of: a time stamp; weather; a temperature; sewage output; a sewage test result; working conditions; and (5) a holiday. The operating condition of the model input data may be determined based on the electric energy consumption data in the production process. And based on the summary and modeling of the production rule, the future working condition, yield and energy consumption are predicted by combining industry scenic index and the like. And simultaneously calculating stock quantity, logistics requirements and the like.
Through the electric energy consumption data of the production process as historical data, the neural network model can be trained to obtain an electric energy consumption prediction model of the production process, and future prediction data can be further obtained according to input.
Specifically, the neural network model may be implemented as: a Recurrent Neural Network (RNN), a feedforward neural network model, a radial basis neural network model, a Long Short Term Memory (LSTM) network model, an Echo State Network (ESN), a gate loop unit (GRU) network model, or a deep residual network model, etc. Preferably, the neural network model is implemented as an LSTM network model. Moreover, the factory can report the output information periodically, and the electric energy consumption prediction model can be updated according to the actual data.
Fig. 4 is an exemplary block diagram of an RNN according to an embodiment of the present invention. The RNN-based power load prediction model is built from multi-dimensional inputs. The multi-dimensional input includes historical electrical load data, time tags, weather data, temperature data, steam consumption data, wastewater production data, and holiday data. When aiming at the problem of time sequence prediction of multi-dimensional input, the circulating neural network has a memory function which is not available in other types of neural networks through a feedback mechanism of the circulating neural network.
Wherein x is 1 ，x 2 ，...，x t Is the input of each step s 1 ，s 2 ，...，s t Is the hidden layer state of each step, y 1 ，y 2 ，...，y t Is the output of each step. A, B, C in fig. 4 are matrices, and are respectively transformation parameters from input to hidden state, hidden state to output, and current state to next state; b h ，b y Is a bias term. They are the content to be learned in training.
Based on the training derived parameters, the forward procedure of RNN can be expressed as:
s t ＝f(A*x t +C*s t-1 +b h )；
y t ＝g(B*s t +b y )；
x t ＝(time stamp，date，temperature t ，weather t ，waste_water_generation t ，holiday，etc.)；
y t ＝(Q t ，Steam t )。
wherein time stamp is a timestamp; date is the date; temperature of t Is the temperature; weather apparatus t Weather is; wasteWater generation t The amount of wastewater is generated; holiday is a holiday.
The process of training to obtain a predictive model of power consumption is described above using RNN as an example, and those skilled in the art will appreciate that this description is exemplary only and is not intended to limit the scope of embodiments of the present invention.
Step 103: and predicting the electric energy consumption data of the workshop in a future production process based on the electric energy consumption prediction model.
Here, based on the prediction data, logistics, warehouse demand, and the like can be calculated.
Wherein: FQ (FQ) k (t) represents predicted future production of plant k; stem (t) represents the amount of Steam produced; s is S k Representing the storage space required for plant k; s is S k_c Representing the process capacity of plant k; FSteam (t) represents the steam demand for future production; FP (Fabry-Perot) k (t) represents a Pre-formThe measured power demand of plant k; FCargo k And (t) represents the predicted plant k logistics demand.
In a specific implementation, the embodiment of the invention mainly comprises:
(1) And acquiring historical data and preprocessing the data, wherein the preprocessing comprises the steps of carrying out complementary processing on the acquired missing data, carrying out correction processing on the data with errors or exceeding the allowable range, and providing the processed data for a load prediction model.
(2) And constructing an electric energy consumption prediction model based on the cyclic neural network, and correcting the model according to the data actually reported by the enterprise.
(3) Performing power consumption prediction based on the power consumption prediction model to obtain FQ 1 (t)，FQ 2 (t)，...，FQ K (t)，FSteam(t)。
The electricity consumption data of the production workshop also comprises parts which do not directly react to production, such as illumination, air conditioning, overhaul and the like. Therefore, in order to analyze the critical production process, it is necessary to decompose the critical loop power consumption data. The key information such as production state, yield, energy efficiency and the like can be calculated through a digital mechanism model of key loop energy consumption and production load. The data of steam consumption, natural gas consumption and the like are often measured at fewer points than the data of electricity consumption, so the granularity is larger. They can reflect production from a certain level, but are less accurate than the power consumption data. Meanwhile, the information such as the wastewater production amount, the laboratory sheet and the like can be used together with the energy consumption data such as steam, electricity and the like to provide a basis for judging production states such as abnormal production judgment. Along with the accumulation of historical data, the embodiment of the invention can also predict energy consumption, yield, logistics demand, inventory, energy consumption demand and the like by a mechanism model, big data processing and other methods, and further provide energy data extension service.
Fig. 2 is an exemplary schematic diagram of a method of predicting plant operation in accordance with an embodiment of the present invention.
In fig. 2, the identification process 20 of the production process includes acquiring the power consumption data 211 of the plant at the first history time, the power consumption data 221 of the plant at the second history time, and the power consumption data 2m1 of the plant at the mth history time. Wherein the number of m is a positive integer greater than 2. Also, the separation process 212 is performed on the consumption data 211, the production evaluation 213 (including determination of the operating condition and the power consumption data) is performed on the basis of the separated data, the separation process 222 is performed on the basis of the consumption data 221, the production evaluation 223 (including determination of the operating condition and the power consumption data) is performed on the basis of the separated data, the separation process 2m2 is performed on the basis of the consumption data 2m1, and the production evaluation 2m3 (including determination of the operating condition and the power consumption data) is performed on the basis of the separated data. Then, the electric power consumption prediction model 40 is trained based on the model input functions of the steam consumption, the sewage yield, the sewage assay result, and the like, and the results of the production evaluations 213, 223, and 2m 3. The power consumption prediction model 40 may then be used to predict the planning strategy for the service 41 of the future generation process. The results of the production evaluations 213, 223 and 2m3 may also be used to determine the service 41 of the current production process.
Fig. 5 is a block diagram of a plant operation prediction apparatus according to an embodiment of the present invention.
As shown in fig. 5, the plant operation prediction apparatus 500 includes:
a separation module 501, configured to separate electric energy consumption data of a workshop in a production process from electric energy consumption data of a workshop in a factory; a determining module 502 for determining an electric energy consumption prediction model adapted to predict electric energy consumption data in a production process based on the electric energy consumption data; a prediction module 503, configured to predict the electrical energy consumption data of the plant in the future production process based on the electrical energy consumption prediction model.
In one embodiment, the separation module 501 is configured to separate the electrical energy consumption data of the workshop in the factory from the electrical energy consumption data of the workshop in the production process based on the energy decomposition mode of the non-parametric factor hidden markov model; or, separating the power consumption data of the workshop in the production process from the power consumption data of the workshop in the factory based on the percentage determined by the historical power distribution data.
In one embodiment, the determining module 502 is further configured to determine an operating condition of the production process based on the electrical energy consumption data of the production process, wherein: when the electric energy consumption data is smaller than a preset first threshold value, determining that the working condition of the production process is a stop condition; when the electric energy consumption data is larger than the first threshold value and smaller than a preset second threshold value, determining that the working condition of the production process is a waiting condition; when the electric energy consumption data is larger than the second threshold value, determining that the working condition of the production process is an active running condition; wherein the second threshold value is greater than the first threshold value.
In one embodiment, the determining module 502 is further configured to determine the output of the production process and the logistics demand of the production process based on the electrical energy consumption data in the production process;
wherein the method comprises the steps of
Wherein the yield of the production process is Q k (t); the logistic requirement of the production process is Cargo k (t); cap is the transport capacity of each transport; p (P) k (t) electrical energy consumption data during production; q (Q) krated Rated production capacity for the production process; p (P) krated Is the rated electric power of the generating process; t is a time parameter.
In one embodiment, the determining module 502 is configured to use the electric energy consumption data in the production process, the steam consumption in the production process, and the model input data together as training data to train the neural network model to obtain an electric energy consumption prediction model; wherein the input data comprises at least one of: a time stamp; weather; a temperature; sewage output; a sewage test result; and (5) a holiday.
In one embodiment, the prediction module 503 is configured to input the input data in the future production process into the electric energy consumption prediction model to predict the electric energy consumption data in the future production process and the steam consumption data in the future production process; determining at least one of the following based on the electrical energy consumption data in the future production process: predicted production of future production processes; predicted logistics requirements for future production processes; predictive storage requirements for future production processes; predicted steam demand for future production processes.
In summary, when the data analysis method provided by the embodiment of the invention is used for analyzing the energy data, a cyclic neural network between multidimensional input and output is established, and data with different characteristics such as power consumption, steam consumption, wastewater production, weather data, holidays and the like are comprehensively considered, so that the future energy load can be comprehensively and effectively predicted by deep learning of production rules among production flows, time sequences of the flows, production rules between production and time, weather, holidays and the like. And through deep learning, circulation and iteration, the model is adjusted by combining with the data actually reported by the enterprise, and the accuracy of the prediction result is ensured. The output prediction data provides key data support for energy consumption management, energy consumption, logistics coordination, metering station use plans, parking lot use plans, storage plans and the like, improves the scheduling accuracy and economy of various aspects of the park, and achieves the purpose of building the intelligent industrial park.
The embodiment of the invention also provides a visual realization device of the car loader with the processor-memory architecture. Fig. 6 is a block diagram of a plant-operated prediction apparatus having a processor-memory architecture according to an embodiment of the present invention.
As shown in fig. 6, the plant operation prediction apparatus 600 includes a processor 601, a memory 602, and a computer program stored in the memory 602 and executable on the processor 601, which when executed by the processor 501, implements the plant operation prediction method as any one of the above.
The memory 602 may be implemented as a variety of storage media such as an electrically erasable programmable read-only memory (EEPROM), a Flash memory (Flash memory), a programmable read-only memory (PROM), and the like. Processor 601 may be implemented to include one or more central processors or one or more field programmable gate arrays that integrate one or more central processor cores. In particular, the central processor or central processor core may be implemented as a CPU or MCU or DSP, etc.
It should be noted that not all the steps and modules in the above processes and the structure diagrams are necessary, and some steps or modules may be omitted according to actual needs. The execution sequence of the steps is not fixed and can be adjusted as required. The division of the modules is merely for convenience of description and the division of functions adopted in the embodiments, and in actual implementation, one module may be implemented by a plurality of modules, and functions of a plurality of modules may be implemented by the same module, and the modules may be located in the same device or different devices.
The hardware modules in the various embodiments may be implemented mechanically or electronically. For example, a hardware module may include specially designed permanent circuits or logic devices (e.g., special purpose processors such as FPGAs or ASICs) for performing certain operations. A hardware module may also include programmable logic devices or circuits (e.g., including a general purpose processor or other programmable processor) temporarily configured by software for performing particular operations. As regards implementation of the hardware modules in a mechanical manner, either by dedicated permanent circuits or by circuits that are temporarily configured (e.g. by software), this may be determined by cost and time considerations.
The foregoing is merely a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (14)
- A method (100) of predicting plant operation, the method (100) comprising:separating power consumption data (101) of a workshop in a production process from power consumption data of the workshop in a factory;determining an electrical energy consumption prediction model (102) adapted to predict electrical energy consumption data in a production process based on the electrical energy consumption data in the production process;electric energy consumption data (103) of the plant in a future production process is predicted based on the electric energy consumption prediction model.
- The method (100) of predicting plant operation according to claim 1, wherein,the separation of the power consumption data (101) of the plant during the production process from the power consumption data of the plant in the plant comprises:separating the electric energy consumption data of the workshop in the production process from the electric energy consumption data of the workshop in the factory based on an energy decomposition mode of a non-parametric factor hidden Markov model; or (b)The power consumption data of the plant during production is separated from the power consumption data of the plant in the plant based on the percentages determined by the historical power distribution data.
- The method (100) of predicting plant operation according to claim 1, further comprising:determining an operating condition of the production process based on the electrical energy consumption data in the production process, wherein:when the electric energy consumption data is smaller than a preset first threshold value, determining that the working condition of the production process is a stop condition;when the electric energy consumption data is larger than the first threshold value and smaller than a preset second threshold value, determining that the working condition of the production process is a waiting condition;when the electric energy consumption data is larger than the second threshold value, determining that the working condition of the production process is an active running condition;wherein the second threshold value is greater than the first threshold value.
- The method (100) of predicting plant operation according to claim 1, further comprising: determining a yield of the production process and a logistics demand of the production process based on the electrical energy consumption data in the production process;wherein the method comprises the steps ofWherein the yield of the production process is Q k (t); the logistic requirement of the production process is Cargo k (t); cap is the transport capacity of each transport; p is p k (t) is electrical energy consumption data in the production process; q (Q) krated Rated capacity for the production process; p (P) krated Is the rated electrical power of the generation process; t is a time parameter.
- The method (100) of predicting plant operation according to claim 1, wherein the determining an electrical energy consumption prediction model (102) adapted to predict electrical energy consumption data in a production process based on electrical energy consumption data in the production process comprises:the electric energy consumption data in the production process, the steam consumption in the production process and the model input data are used as training data together to train the neural network model so as to obtain the electric energy consumption prediction model; wherein the model input data comprises at least one of: a time stamp; weather; a temperature; sewage output; a sewage test result; working conditions; and (5) a holiday.
- The method (100) of predicting plant operation according to claim 5, wherein predicting electrical energy consumption data (103) of the plant in a future production process based on the electrical energy consumption prediction model comprises:inputting model input data in a future production process into the electric energy consumption prediction model to predict electric energy consumption data in the future production process and steam consumption data in the future production process; the method further comprises the steps of:determining at least one of the following based on the electrical energy consumption data in the future production process: predicted production of future production processes; predicted logistics requirements for future production processes; predicted storage requirements for future production processes.
- A plant operation prediction apparatus (500), comprising:a separation module (501) for separating the power consumption data of the workshop in the production process from the power consumption data of the workshop in the factory;a determining module (502) for determining an electrical energy consumption prediction model adapted to predict electrical energy consumption data in a production process based on the electrical energy consumption data in the production process;a prediction module (503) for predicting electrical energy consumption data of the plant in a future production process based on the electrical energy consumption prediction model.
- The plant operation prediction device (500) according to claim 7, wherein,the separation module (501) is used for separating the electric energy consumption data of the workshop in the production process from the electric energy consumption data of the workshop in the factory based on the energy decomposition mode of the non-parametric factor hidden Markov model; or, based on the percentage determined by the historical power distribution data, separating the power consumption data of the plant during the production process from the power consumption data of the plant in the plant.
- The plant operation prediction device (500) according to claim 7, wherein,the determining module (502) is further configured to determine an operating condition of the production process based on the electrical energy consumption data in the production process, wherein: when the electric energy consumption data is smaller than a preset first threshold value, determining that the working condition of the production process is a stop condition; when the electric energy consumption data is larger than the first threshold value and smaller than a preset second threshold value, determining that the working condition of the production process is a waiting condition; when the electric energy consumption data is larger than the second threshold value, determining that the working condition of the production process is an active running condition; wherein the second threshold value is greater than the first threshold value.
- The plant operation prediction device (500) according to claim 7, wherein,-the determining module (502) is further configured to determine a yield of the production process and a logistics requirement of the production process based on the electrical energy consumption data in the production process;wherein the method comprises the steps ofWherein the yield of the production process is Q k (t); the logistic requirement of the production process is Cargo k (t); cap is the transport capacity of each transport; p (P) k (t) is electrical energy consumption data in the production process; q (Q) krated Rated capacity for the production process; p (P) krated Is the rated electrical power of the generation process; t is a time parameter.
- The plant operation prediction device (500) according to claim 7, wherein,the determining module (502) is configured to use the electric energy consumption data in the production process, the steam consumption in the production process and the model input data together as training data to train a neural network model, so as to obtain the electric energy consumption prediction model; wherein the model input data comprises at least one of: a time stamp; weather; a temperature; sewage output; a sewage test result; and (5) a holiday.
- The plant operation prediction device (500) according to claim 11, wherein,the prediction module (503) is configured to input model input data in a future production process into the electric energy consumption prediction model to predict electric energy consumption data in the future production process and steam consumption data in the future production process; determining at least one of the following based on the electrical energy consumption data in the future production process: predicted production of future production processes; predicted logistics requirements for future production processes; predicted storage requirements for future production processes.
- A plant-operated prediction apparatus (600), characterized by comprising a processor (601) and a memory (602);the memory (602) has stored therein an application executable by the processor (601) for causing the processor (601) to perform the plant operation prediction method (100) according to any one of claims 1 to 6.
- A computer readable storage medium, characterized in that computer readable instructions are stored therein for performing the prediction method (100) of a plant operation according to any of the claims 1 to 6.
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2021/115691 WO2023028842A1 (en) | 2021-08-31 | 2021-08-31 | Factory operation prediction method and apparatus, and computer-readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117751373A true CN117751373A (en) | 2024-03-22 |
Family
ID=85410725
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202180101142.5A Pending CN117751373A (en) | 2021-08-31 | 2021-08-31 | Method and device for predicting factory operation and computer readable storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN117751373A (en) |
| WO (1) | WO2023028842A1 (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116755343B (en) * | 2023-08-18 | 2023-12-19 | 兆和能源(威海)有限公司 | Self-learning fuzzy control-based electricity economizer |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2013078541A1 (en) * | 2011-11-29 | 2013-06-06 | Energy Aware Technology Inc. | Method and system for forecasting power requirements using granular metrics |
| JP6602609B2 (en) * | 2015-09-04 | 2019-11-06 | パナソニック インテレクチュアル プロパティ コーポレーション オブ アメリカ | Power usage state estimation method |
| CN109344168A (en) * | 2018-10-29 | 2019-02-15 | 上海新增鼎数据科技有限公司 | Supervise method, platform, equipment and the storage medium of plant produced operation state |
| CN109709912B (en) * | 2018-12-20 | 2021-04-20 | 广西程天电子科技有限公司 | Energy management control method and system based on Internet of things |
| CN111783947A (en) * | 2020-06-24 | 2020-10-16 | 上海凯营新能源技术有限公司 | Energy consumption prediction method based on LSTM neural network |
| CN111861206A (en) * | 2020-07-20 | 2020-10-30 | 国网上海市电力公司 | Industrial industry prosperity index obtaining method based on enterprise electric power big data |
| CN111950794A (en) * | 2020-08-18 | 2020-11-17 | 上海仪电（集团）有限公司中央研究院 | Park energy consumption prediction method, system, equipment and storage medium |
| CN112711229B (en) * | 2020-12-09 | 2022-06-17 | 万洲电气股份有限公司 | Intelligent optimization energy-saving system based on multi-correlation factor energy consumption prediction |
| CN112990712A (en) * | 2021-03-19 | 2021-06-18 | 成都青云之上信息科技有限公司 | Enterprise production and operation analysis method and system based on power consumption monitoring |
-
2021
- 2021-08-31 CN CN202180101142.5A patent/CN117751373A/en active Pending
- 2021-08-31 WO PCT/CN2021/115691 patent/WO2023028842A1/en active Application Filing
Also Published As
| Publication number | Publication date |
|---|---|
| WO2023028842A1 (en) | 2023-03-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| He et al. | Joint decision-making of parallel machine scheduling restricted in job-machine release time and preventive maintenance with remaining useful life constraints | |
| Mathew et al. | Regression kernel for prognostics with support vector machines | |
| CN111695744B (en) | Maintenance equipment demand prediction analysis system based on big data | |
| CN114595623A (en) | XGboost algorithm-based unit equipment reference value prediction method and system | |
| CN116843071B (en) | Transportation network operation index prediction method and device for intelligent port | |
| CN117541197B (en) | Intelligent building method and system based on BIM and AIOT data driving | |
| Zhao et al. | Equipment fault forecasting based on ARMA model | |
| CN114169254A (en) | Abnormal energy consumption diagnosis method and system based on short-term building energy consumption prediction model | |
| Baranov et al. | SMART technologies for transport tests networks, exploitation and repair tools | |
| Harnischmacher et al. | Two-sided sustainability: Simulating battery degradation in vehicle to grid applications within autonomous electric port transportation | |
| CN117751373A (en) | Method and device for predicting factory operation and computer readable storage medium | |
| Li et al. | Computational logistics for container terminal handling systems with deep learning | |
| Gaidar et al. | Mathematical method for optimising the transport and logistics industry | |
| Kim et al. | Forecasting future electric power consumption in Busan New Port using a deep learning model | |
| CN105404940B (en) | Maintenance resource prediction method for ship use stage | |
| Karunakaran et al. | Toward evolving dispatching rules for dynamic job shop scheduling under uncertainty | |
| Sun et al. | Application of the LP-ELM model on transportation system lifetime optimization | |
| Chen et al. | Short-term traffic flow prediction with recurrent mixture density network | |
| CN116882673A (en) | Coal supply chain system and scheduling method | |
| CN111027760A (en) | Power load prediction method based on least square vector machine | |
| Tu et al. | Estimation of machine parameters in exponential serial lines using feedforward neural networks | |
| Zhang | Revolutionizing Digital Marketing Industrial Internet of Things-enabled Supply Chain Management in Smart Manufacturing | |
| Bai et al. | A systemic method of traffic flow velocity prediction in narrow waterways using ais data | |
| Petrouš et al. | Modeling of mixed data for Poisson prediction | |
| Candelieri et al. | Data efficient learning of implicit control strategies in Water Distribution Networks |
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 |