WO2023028842A1 - Factory operation prediction method and apparatus, and computer-readable storage medium - Google Patents

Abstract

Disclosed in the embodiments of the present invention are a factory operation prediction method and apparatus, and a computer-readable storage medium. The method comprises: separating, from electric energy consumption data of a workshop in a factory, electric energy consumption data of the workshop during a production process; on the basis of the electric energy consumption data during the production process, determining an electric energy consumption prediction model which matches the prediction of electric energy consumption data during a production process; and on the basis of the electric energy consumption prediction model, predicting electric energy consumption data of the workshop during a future production process. For a factory where data, such as a production situation and production prediction, is not provided, the operation situation of the factory can be predicted on the basis of electric energy consumption data during a production process, thereby facilitating the provision of a plurality of services such as an energy service or a logistics service.

Description

Prediction method, device, and computer-readable storage medium for factory operation technical field

The invention relates to the technical field of factory data processing, in particular to a method, device and computer-readable storage medium for predicting factory operations.

Background technique

The construction of smart industrial park is an important research field that has attracted much attention in the operation and management of modern industrial parks, and its technology is expanding and maturing day by day. With the development of technology, there is an increasing demand for developing smart industrial parks in various places, so as to provide better services for the enterprises in the parks. The construction of a smart park includes energy supply, transportation and logistics, security, environmental protection, and industrial collaboration. The intelligence of these sectors usually requires the support of the production status and production plan data of each enterprise. However, in reality, companies usually do not provide data such as production status and production forecasts for commercial confidentiality or other purposes.

At present, there is still a lack of targeted and efficient accurate methods and models for issues such as production status judgment, energy services, logistics services, and metering station usage plans under the large scenario of lack of production data in park enterprises, and it is difficult to meet the needs of smart park construction. The need to obtain accurate enterprise production data.

Contents of the invention

Embodiments of the present invention provide a method, device, and computer-readable storage medium for predicting factory operations.

A method of forecasting plant operations, the method comprising:

Separate 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 the electric energy consumption data in the production process based on the electric energy consumption data in the production process;

The electric energy consumption data of the workshop in the future production process is predicted based on the electric energy consumption prediction model.

It can be seen that for factories that lack enterprise production data, the power consumption data in the future production process can be predicted based on the power consumption data in the production process, so as to facilitate the provision of energy services.

In one embodiment, the 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 includes:

Based on the energy decomposition method of the non-parametric factor hidden Markov model, the electric energy consumption data of the workshop in the production process is separated from the electric energy consumption data of the workshop in the factory; or

Based on the percentage determined from the historical power distribution data, the power consumption data of the workshop in the factory during the production process is separated from the power consumption data of the workshop.

Therefore, by isolating the data of electrical energy consumption during the production process, it is possible to more accurately predict the condition of the plant operation.

In one embodiment, also include:

Determining the working status of the production process based on the electric energy consumption data in the production process, wherein:

When the power consumption data is less than a preset first threshold value, it is determined that the working state of the production process is a stop state;

When the power consumption data is greater than the first threshold and less than a preset second threshold, it is determined that the working state of the production process is a waiting state;

When the power consumption data is greater than the second threshold value, it is determined that the working state of the production process is an active running state;

Wherein the second threshold value is greater than the first threshold value.

It can be seen that the working status of the production process can be determined based on the electric energy consumption data of the production process, so as to facilitate understanding of the working status.

In one embodiment, further comprising: determining the output of the production process and the logistics demand of the production process based on the electric energy consumption data in the production process;

in

Wherein the output of the production process is Qk (t); the logistics demand of the production process is Cargo _k (t); Cap is the transportation capacity of each vehicle; P _k (t) is the electric energy in the production process Consumption data; Q _krated is the rated production capacity of the production process; P _krated is the rated electric power of the generation process; t is the time parameter.

Therefore, based on the power consumption data of the production process, the output and logistics requirements can be determined to facilitate the provision of logistics services.

In one embodiment, the determining an electric energy consumption prediction model adapted to predict the electric energy consumption data in the production process based on the electric energy consumption data in the production process includes:

The electric energy consumption data in the production process, the steam consumption in the production process and the model input data are jointly used as training data to train the neural network model to obtain the electric energy consumption prediction model; wherein the model input data includes the following At least one of: time stamp; weather; temperature; sewage production; sewage assay results; work status; holidays.

Therefore, an electric energy consumption prediction model can be established based on artificial intelligence technology. In particular, the trained neural network model is more accurate by using the determined working conditions as model input data.

In one embodiment, the prediction of the electric energy consumption data of the workshop in the future production process based on the electric energy consumption prediction model includes:

Inputting the model 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; the method also includes:

At least one of the following is determined based on the electrical energy consumption data in the future production process: predicted output of the future production process; predicted logistics requirements of the future production process; predicted storage requirements of the future production process.

It can be seen that predicting the power consumption data in the future production process based on the power consumption prediction model is convenient for planning the logistics services and storage services that need to be provided in the future. Moreover, based on the predicted steam consumption data in the future production process, it is convenient to plan the steam service that needs to be provided in the future.

A plant operation forecasting device, comprising:

The separation module is used to separate 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, configured to determine an electric energy consumption prediction model adapted to predict the electric energy consumption data in the production process based on the electric energy consumption data in the production process;

A prediction module, configured to predict 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 that lack enterprise production data, the power consumption data in the future production process can be predicted based on the power consumption data in the production process, so as to facilitate the provision of energy services.

In one embodiment, the separation module is used to separate 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 energy decomposition method 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 workshop in the production process from the power consumption data of the workshop in the factory.

Therefore, by isolating the data of electrical energy consumption during the production process, it is possible to more accurately predict the condition of the plant operation.

In one embodiment, the determination module is further configured to determine the working status of the production process based on the power consumption data in the production process, wherein: when the power consumption data is less than a preset first threshold value, it is determined that the working condition of the production process is a stop condition; when the power consumption data is greater than the first threshold value and less than a preset second threshold value, it is determined that the working condition of the production process It is a waiting state; when the power consumption data is greater than the second threshold value, it is determined that the working state of the production process is an active running state; wherein the second threshold value is greater than the first threshold value.

It can be seen that the working status of the production process can be determined based on the electric energy consumption data of the production process, so as to facilitate understanding of the working status.

In one embodiment, the determination module is further configured to determine the output of the production process and the logistics demand of the production process based on the electric energy consumption data in the production process;

in

Wherein the output of the production process is Q _k (t); the logistics demand of the production process is Cargo _k (t); Cap is the transportation capacity of each vehicle; P _k (t) is the Electric energy consumption data; Q _krated is the rated production capacity of the production process; P _krated is the rated electric power of the generation process; t is a time parameter.

Therefore, based on the power consumption data of the production process, the output and logistics requirements can be determined to facilitate the provision of logistics services.

In one embodiment, the determination module is configured to use the electric energy consumption data in the production process, the steam consumption in the production process and 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 includes at least one of the following: time stamp; weather; temperature; sewage production; sewage test results; holidays.

Therefore, an electric energy consumption prediction model can be established based on artificial intelligence technology. In particular, the trained neural network model is more accurate by using the determined working conditions as model input data.

In one embodiment, the prediction module is configured to input the model input data in the future production process into the electric energy consumption prediction model, so as to predict the electric energy consumption data in the future production process and the steam in the future production process Consumption data; determining at least one of the following based on the electrical energy consumption data in the future production process: predicted output of the future production process; predicted logistics requirements of the future production process; predicted storage requirements of the future production process.

It can be seen that predicting the power consumption data in the future production process based on the power consumption prediction model is convenient for planning the logistics services and storage services that need to be provided in the future. Moreover, based on the predicted steam consumption data in the future production process, it is convenient to plan the steam service that needs to be provided in the future.

A plant operation forecasting device comprising a processor and a memory;

An application program that can be executed by the processor is stored in the memory, which is used to make the processor execute any method for predicting plant operation as described above.

A computer-readable storage medium, in which computer-readable instructions are stored, and the computer-readable instructions are used to execute any one of the methods for predicting plant operation as described above.

Description of drawings

Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, so that those of ordinary skill in the art will be more aware of the above-mentioned and other features and advantages of the present invention, in the accompanying drawings:

FIG. 1 is a flowchart of a method for predicting plant operation according to an embodiment of the present invention.

Fig. 2 is an exemplary schematic diagram of a method for predicting plant operation according to an embodiment of the present invention.

FIG. 3 is an exemplary schematic diagram of a non-parametric factorial hidden Markov model (NFHMM) process according to an embodiment of the present invention.

FIG. 4 is an exemplary structure diagram of a recurrent neural network (RNN) according to an embodiment of the present invention.

Fig. 5 is a configuration diagram of a plant operation prediction device according to an embodiment of the present invention.

FIG. 6 is a structural diagram of a plant operation forecasting device with a processor-memory architecture according to an embodiment of the present invention.

Wherein, the reference signs are as follows:

label	meaning
100	Predictive Methods for Plant Operations
101～103	step
20	Identification process of the production process
211, 221, 2m1	Electric energy consumption data of the workshop
212, 222, 2m2			separation process
213, 223, 2m3		production evaluation
40	Electric Energy Consumption Prediction Model
41	Serve
31	Steam consumption, effluent production, effluent assay results, etc.
32	Production process statistics provided by the factory
50	prior distribution
51	hyperparameters
52	Transformation Matrix Parameters

53	device status
54	Equipment energy consumption characteristics
55	Energy measurement
500	Prediction device for factory operation
501	Separation module
502	Determine the module
503	prediction module
600	Prediction device for factory operation
601	processor
602	memory

Detailed ways

In order to make the purpose, technical solution and advantages of the present invention clearer, the following examples are given to further describe the present invention in detail.

For the sake of brevity and intuition in description, the solution of the present invention is described below by describing several representative implementation manners. Numerous details in the embodiments are only used to help the understanding of the solutions of the present invention. But obviously, the technical solutions of the present invention may not be limited to these details when implemented. In order to avoid unnecessarily obscuring the solution of the present invention, some embodiments are not described in detail, but only a framework is given. Hereinafter, "including" means "including but not limited to", and "according to..." means "at least according to, but not limited to only based on...". Due to the language habits of Chinese, when the quantity of a component is not specifically indicated below, it means that the component can be one or more, or can be understood as at least one.

Considering the current production status judgment, energy service, logistics service, metering station usage plan and other issues in the big scenario of lack of production status of park enterprises, there is still a lack of targeted and efficient accurate methods and models, and it is difficult to meet the needs of smart parks. Based on the current situation of the demand for obtaining accurate enterprise production data, the embodiment of the present invention proposes a method to estimate the production status of the enterprise based on the power consumption of the key production process of the key workshop in the factory based on the decomposition of energy consumption, and apply it to the smart industry The construction of the park solves the problem of lack of key data faced by energy services and logistics services in the construction of smart industrial parks.

FIG. 1 is a flowchart of a method for predicting plant operation according to an embodiment of the present invention.

As shown in Figure 1, the method 100 includes:

Step 101: Separate the power consumption data of the workshop in the production process from the power consumption data of the workshop in the factory.

In reality, the industrial park often only measures the power consumption of each production workshop (or section), but does not have detailed power consumption data of each equipment in each enterprise workshop. At the same time, according to the power consumption of each production workshop, there may also be a situation where two or more workshops share one measurement point.

First, based on the professional knowledge of the factory's production characteristics, it is determined that the production workshop with key functions and that can effectively collect energy consumption data is the "workshop" described in step 101. The power consumption data of the workshop mentioned in step 101 usually also includes parts that do not directly respond to production, such as lighting, air conditioning, and maintenance. Therefore, it is necessary to separate the power consumption data of the parts that do not directly respond to production in step 101 .

In one embodiment, based on the non-parametric factorial hidden Markov model (NFHMM) energy decomposition method, the power consumption data of the workshop in the production process is separated from the power consumption data of the workshop in the factory. Hidden Markov model is a kind of Markov chain. Its state cannot be observed directly, but it can be observed through the observation vector sequence. Each observation vector is expressed as various states through certain probability density distributions. Each An observation vector is generated from a sequence of states with a corresponding probability density distribution. So, the Hidden Markov Model is a double stochastic process with a hidden Markov chain with a certain number of states and a set of explicit random functions.

The energy decomposition method of NFHMM is described in detail below. FIG. 3 is an exemplary schematic diagram of NFHMM processing according to an embodiment of the present invention.

Assume that the aggregate energy consumption signal Y(t) of K devices (or processes) in the workshop of the factory is obtained, where t is from 1 to T. Among them, the K equipment includes the equipment of the generation process, and also includes the equipment that does not directly reflect the production such as lighting, air conditioning, and maintenance. Here, no prior knowledge 50 about the number of devices is assumed, ie K is unknown. Let Z be a T*K matrix, and the z _{t corresponding to the (t, k) item, k} represents the state of device k at time t. In Figure 3, NFHMM is described as follows:

Assume that the state of each device evolves according to the transition matrix parameter 52:

;in

The state z _{t of the k-th device at time t, k} follows

Put the prior distribution 50 on μ _k ~ Beta(α/K, 1), b _k ~ Beta(γ, δ). Hyperparameters 51 include α, γ, δ, and H. The device status 53 includes the status of each device. The device energy consumption feature 54 is Θ, where the kth entry of the vector represents the Gaussian distributed power of the kth device. The observed power measurement 55 (mixed signal y) is generated by the emission model Y=ZΘ+ε, where ε is the measurement noise.

The above exemplarily describes a typical example of using NFHMM to process the power consumption data of the separation workshop during the production process. Those skilled in the art can realize that this description is only exemplary and is not used to limit the scope of protection of the embodiments of the present invention. .

In one embodiment, based on the percentage determined from the historical power distribution data, the power consumption data of the workshop in the factory during the production process is separated from the power consumption data of the workshop.

In one embodiment, the working status of the production process is determined based on the electric energy consumption data in the production process, wherein: when the electric energy consumption data is less than a preset first threshold value, the production process is determined to be The working state of the production process is a stop state; when the power consumption data is greater than the first threshold value and less than the preset second threshold value, it is determined that the working state of the production process is a waiting state; when the electric energy When the consumption data is greater than the second threshold value, it is determined that the working state of the production process is an active running state; wherein the second threshold value is greater than the first threshold value.

Specifically, C _k (t)=Condition_map(P _k (t));

C _k (t) is the working condition. An exemplary matching method is as follows, and other matching methods can also be used: if P _k <P ₁ , the workshop is in shutdown; if P ₁ <P _k <P ₂ , the workshop is on standby; if P ₂ <P _k , the workshop in processing.

Moreover, according to the production conditions of each workshop, logistics, etc. can be reasonably planned. Moreover, based on the decomposed power consumption and the relationship between industry power consumption and production, the output of the enterprise can be calculated. Based on the obtained power consumption of the equipment and the possible operating conditions of the equipment, the power consumption can be matched to the operating mode of the equipment. Based on the working conditions and duration of each link, the production law of the enterprise can be obtained. Excavate the relationship between key processes and their significance to energy consumption: convert consumption data and the duration of each process into the output of various intermediate products, warehouse stock, and corresponding logistics needs, etc.

In one embodiment, the method further includes: determining the output of the production process and the logistics demand of the production process based on the electric energy consumption data in the production process;

in

Wherein the output of the production process is Q _k (t); the logistics demand of the production process is Cargo _k (t); Cap is the transportation capacity of each vehicle; P _k (t) is the Electric energy consumption data; Q _krated is the rated production capacity of the production process; t is the time parameter.

Step 102: Determine an electric energy consumption prediction model adapted to predict the electric energy consumption data in the production process based on the electric energy consumption data.

Here, the electric energy consumption data in the production process, the steam consumption in the production process and the model input data are jointly used as training data to train the neural network model to obtain the electric energy consumption prediction model; wherein the model input data includes the following At least one of: time stamp; weather; temperature; effluent production; effluent assay results; work status; holidays. Wherein, the working status as the input data of the model may be determined based on the electric energy consumption data in the production process. Moreover, based on the summary and modeling of production laws, combined with the industry prosperity index, etc., the future working conditions, output and energy consumption are predicted. Simultaneously calculate inventory, logistics demand, etc.

Through the electric energy consumption data of the production process as historical data, the neural network model can be trained to obtain the electric energy consumption prediction model of the production process, and then the future prediction data can be obtained according to the input.

Specifically, the neural network model can be implemented as: recurrent neural network (RNN), feedforward neural network model, radial basis neural network model, long short-term memory (LSTM) network model, echo state network (ESN), gate recurrent unit ( GRU) network model or deep residual network model, etc. Preferably, the neural network model is implemented as an LSTM network model. Moreover, the factory will regularly report production information, and the power consumption prediction model can be updated based on actual data.

Fig. 4 is an exemplary structural diagram of an RNN in an embodiment of the present invention. The power load forecasting model based on RNN is established by multi-dimensional input. Multidimensional inputs include historical power load data, time stamps, weather data, temperature data, steam consumption data, wastewater production data, and holiday data. For the time series prediction problem of multi-dimensional input, the recurrent neural network has a memory function that other types of neural networks do not have through its own feedback mechanism.

Among them, x 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 , ... , _yt is the output of each step. A, B, and C in Figure 4 are all matrices, which are transformation parameters from input to hidden layer state, hidden layer state to output, and current state to next state; b _h and b _y are bias items. They are what are to be learned in training.

Based on the parameters obtained by training, the forward process 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 ).

Among them, time stamp is the timestamp; date is the date; temperature _t is the temperature; weather _t is the weather; waste_water_generation _t is the amount of wastewater generated; holiday is the holiday.

The above uses RNN as an example to describe the process of training and obtaining the electric energy consumption prediction model. Those skilled in the art can realize that this description is only exemplary and is not intended to limit the scope of protection of the embodiments of the present invention.

Step 103: Predict the electric energy consumption data of the workshop in the future production process based on the electric energy consumption prediction model.

Here, based on forecast data, logistics, warehousing needs, etc. can be calculated.

Among them: FQ _k (t) represents the predicted future output of workshop k; Steam(t) represents the amount of steam produced; S _k represents the storage space required by workshop k; S _{k_c} represents the processing capacity of workshop k; FSteam(t) represents the steam demand for future production; FP _k (t) represents the predicted power demand of workshop k; FCargo _k (t) represents the predicted logistics demand of workshop k.

In specific implementation, the embodiment of the present invention mainly includes:

(1) Obtain historical data and perform data preprocessing. Preprocessing includes supplementary processing of missing data, correction of data that is incorrect or exceeds the allowable range, and provides the processed data to the load forecasting model. .

(2) Construct a prediction model of electric energy consumption based on the cyclic neural network, and correct the model according to the data actually reported by the enterprise.

(3) Predict the power consumption based on the power consumption prediction model to obtain FQ ₁ (t), FQ ₂ (t), . . . , FQ _K (t), FSteam(t).

The power consumption data of the production workshop also includes lighting, air conditioning, maintenance and other parts that do not directly reflect production. Therefore, in order to analyze the key production process, it is necessary to decompose the key circuit power consumption data. Through the digital mechanism model of key circuit energy consumption and production load, key information such as production status, output, and energy efficiency can be calculated. Data such as steam consumption and natural gas consumption often have fewer measurement points than electricity consumption data, so the granularity is larger. They reflect production somewhat, but not as accurately as electricity consumption data. At the same time, information such as the amount of waste water produced and test sheets can be used together with energy consumption data such as steam and electricity to provide a basis for judging production status, such as judging abnormal production. With the accumulation of historical data, through methods such as mechanism models and big data processing, the embodiments of the present invention can also predict energy consumption, output, logistics demand, inventory, energy consumption demand, etc., and further provide energy data extension services.

Fig. 2 is an exemplary schematic diagram of a method for predicting plant operation according to an embodiment of the present invention.

In Fig. 2, the identification process 20 of the production process includes obtaining the power consumption data 211 of the workshop at the first historical time, the power consumption data 221 of the workshop at the second historical time, and the power consumption data of the workshop at the mth historical time Data 2m1. Wherein the number of m is a positive integer greater than 2. Furthermore, a separation process 212 is performed on the consumption data 211, a production assessment 213 is performed based on the separated data (including determining operating conditions and power consumption data), a separation process 222 is performed on the consumption data 221, a production assessment 223 is performed based on the separated data ( Including determining working conditions and power consumption data), performing a separation process 2m2 on consumption data 2m1, performing production evaluation 2m3 based on the separated data (including determining working conditions and power consumption data). Then, based on the model input functions such as steam consumption, sewage production, sewage test results and the results of production evaluation 213, 223 and 2m3, the electric energy consumption prediction model 40 is trained. Then, the planning strategy of the service 41 of the future generation process can be predicted by using the electric energy consumption prediction model 40 . The service 41 of the current production process can also be determined using the results of the production evaluation 213, 223 and 2m3.

Fig. 5 is a configuration diagram of a plant operation prediction device according to an embodiment of the present invention.

As shown in Figure 5, the plant operation forecasting device 500 includes:

The separation module 501 is used to separate the power consumption data of the workshop in the production process from the power consumption data of the workshop in the factory; the determination module 502 is used to determine the power consumption data suitable for predicting the production process based on the power consumption data. Electric energy consumption prediction model; prediction module 503, used to predict the electric energy consumption data of the workshop in the future production process based on the electric energy consumption prediction model.

In one embodiment, the separation module 501 is used to separate 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 energy decomposition method of the non-parametric factor hidden Markov model; or, Based on the percentage determined from the historical power distribution data, the power consumption data of the workshop in the production process is separated from the power consumption data of the workshop in the factory.

In one embodiment, the determining module 502 is further configured to determine the working status of the production process based on the power consumption data in the production process, wherein: when the power consumption data is less than the preset first threshold value, determine the operating status of the production process The working state is a stop state; when the power consumption data is greater than the first threshold value and less than the preset second threshold value, it is determined that the working state of the production process is a waiting state; when the power consumption data is greater than the second threshold value , determining that the working state of the production process is an active running state; wherein the second threshold value is greater than the first threshold value.

In one embodiment, the determination module 502 is also used to determine the output of the production process and the logistics demand of the production process based on the electric energy consumption data in the production process;

in

The output of the production process is Q _k (t); the logistics demand of the production process is Cargo _k (t); Cap is the transportation capacity of each vehicle; P _k (t) is the power consumption data in the production process; Q _krated is the rated production capacity of the production process; P _krated is the rated electric power of the generation process; t is the time parameter.

In one embodiment, the determination module 502 is used to use the data of electric energy consumption in the production process, the steam consumption in the production process, and the input data of the model together as training data to train the neural network model, so as to obtain the prediction model of electric energy consumption; Wherein the input data includes at least one of the following: time stamp; weather; temperature; sewage production; sewage test results; holidays.

In one embodiment, the prediction module 503 is used to input the input data in the future production process into the electric energy consumption prediction model, so as to predict the electric energy consumption data in the future production process and the steam consumption data in the future production process; The electrical energy consumption data in the process determines at least one of: predicted production capacity of the future production process; predicted logistics requirements of the future production process; predicted storage needs of the future production process; predicted steam demand of the future production process.

To sum up, the data analysis method proposed in the embodiment of the present invention adopts a recurrent neural network between multi-dimensional input and output when analyzing energy data, and comprehensively considers power consumption, steam consumption, waste water production, weather data, Data with different characteristics such as holidays, deep learning between various production processes, between the time series of the process itself, and between production and time, weather, and holidays, etc., to comprehensively and effectively predict future energy loads. After deep learning, circulation and iteration, and combined with the data actually reported by the enterprise, the model is adjusted to ensure the accuracy of the forecast results. The output forecast data provides key data support for energy consumption management, energy consumption, logistics coordination, metering station usage plan, parking lot usage plan, storage plan, etc., improves the accuracy and economy of all aspects of the park's scheduling, and achieves the construction of a smart industrial park the goal of.

The embodiment of the present invention also proposes a vision realization device for a vehicle loading machine with a processor-memory architecture. FIG. 6 is a structural diagram of a plant operation forecasting device with a processor-memory architecture according to an embodiment of the present invention.

As shown in Figure 6, the plant operation forecasting device 600 includes a processor 601, a memory 602, and a computer program stored on the memory 602 and operable on the processor 601. When the computer program is executed by the processor 501, any of the above A forecasting method for plant operations.

Wherein, the memory 602 can be specifically implemented as various storage media such as electrically erasable programmable read-only memory (EEPROM), flash memory (Flash memory), and programmable program read-only memory (PROM). The processor 601 may be implemented to include one or more central processing units or one or more field programmable gate arrays, wherein the field programmable gate arrays integrate one or more central processing unit cores. Specifically, the central processing unit or central processing unit core may be implemented as a CPU or MCU or DSP, and so on.

It should be noted that not all steps and modules in the above-mentioned processes and structure diagrams are necessary, and some steps or modules can be ignored according to actual needs. The execution order of each step is not fixed and can be adjusted as needed. The division of each module is only to facilitate the description of the functional division adopted. In actual implementation, one module can be divided into multiple modules, and the functions of multiple modules can also be realized by the same module. These modules can be located in the same device. , or on a different device.

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 (such as special-purpose processors, such as FPGAs or ASICs) to perform specific operations. Hardware modules may also include programmable logic devices or circuits (eg, including general-purpose processors or other programmable processors) temporarily configured by software to perform particular operations. As for implementing the hardware module in a mechanical way, using a dedicated permanent circuit, or using a temporarily configured circuit (such as configured by software) to realize the hardware module, it can be decided according to cost and time considerations.

The above are only preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims (14)

1. A method (100) for forecasting factory operations, characterized in that the method (100) comprises:

   Separate the power consumption data of the workshop in the production process from the power consumption data of the workshop in the factory (101);

   Based on the electric energy consumption data in the production process, determining an electric energy consumption prediction model adapted to predict the electric energy consumption data in the production process (102);

   Predict the electric energy consumption data of the workshop in the future production process based on the electric energy consumption prediction model (103).
2. The prediction method (100) of factory operation according to claim 1, it is characterized in that,

   The 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 (101) includes:

   Based on the energy decomposition method of the non-parametric factor hidden Markov model, the electric energy consumption data of the workshop in the production process is separated from the electric energy consumption data of the workshop in the factory; or

   Based on the percentage determined from the historical power distribution data, the power consumption data of the workshop in the factory during the production process is separated from the power consumption data of the workshop.
3. The prediction method (100) of factory operation according to claim 1, is characterized in that, also comprises:

   Determining the working status of the production process based on the electric energy consumption data in the production process, wherein:

   When the power consumption data is less than a preset first threshold value, it is determined that the working state of the production process is a stop state;

   When the power consumption data is greater than the first threshold and less than a preset second threshold, it is determined that the working state of the production process is a waiting state;

   When the power consumption data is greater than the second threshold value, it is determined that the working state of the production process is an active running state;

   Wherein the second threshold value is greater than the first threshold value.
4. The method (100) for predicting factory operation according to claim 1, further comprising: determining the output of the production process and the logistics demand of the production process based on the electric energy consumption data in the production process;

   in

   

   Wherein the output of the production process is Q _k (t); the logistics demand of the production process is Cargo _k (t); Cap is the transportation capacity of each transport tool; p _k (t) is in the production process Electric energy consumption data; Q _krated is the rated production capacity of the production process; P _krated is the rated electric power of the generation process; t is a time parameter.
5. The prediction method (100) of factory operation according to claim 1, characterized in that, based on the electric energy consumption data in the production process, determining an electric energy consumption prediction model (102) adapted to predict the electric energy consumption data in the production process )include:

   The electric energy consumption data in the production process, the steam consumption in the production process and the model input data are jointly used as training data to train the neural network model to obtain the electric energy consumption prediction model; wherein the model input data includes the following At least one of: time stamp; weather; temperature; sewage production; sewage assay results; work status; holidays.
6. The prediction method (100) for factory operation according to claim 5, wherein the prediction of the electric energy consumption data (103) of the workshop in the future production process based on the electric energy consumption prediction model comprises:

   Inputting the model 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; the method also includes:

   At least one of the following is determined based on the electrical energy consumption data in the future production process: predicted output of the future production process; predicted logistics requirements of the future production process; predicted storage requirements of the future production process.
7. A plant operation forecasting device (500), characterized in that it comprises:

   A separation module (501), configured to separate 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), configured to determine an electric energy consumption prediction model adapted to predict the electric energy consumption data in the production process based on the electric energy consumption data in the production process;

   A prediction module (503), configured to predict the electric energy consumption data of the workshop in the future production process based on the electric energy consumption prediction model.
8. The prediction device (500) of factory operation according to claim 7, it is characterized in that,

   The separation module (501) is used to separate 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 energy decomposition method of the non-parametric factor hidden Markov model; or, Based on the percentage determined from the historical power distribution data, the power consumption data of the workshop in the factory during the production process is separated from the power consumption data of the workshop.
9. The prediction device (500) of factory operation according to claim 7, it is characterized in that,

   The determination module (502) is further configured to determine the working status of the production process based on the electric energy consumption data in the production process, wherein: when the electric energy consumption data is less than a preset first threshold value, Determining that the working state of the production process is a stop state; when the power consumption data is greater than the first threshold value and less than a preset second threshold value, determining that the working state of the production process is a waiting state ; When the electric energy consumption data is greater than the second threshold value, determine that the working state of the production process is an active running state; wherein the second threshold value is greater than the first threshold value.
10. The prediction device (500) of factory operation according to claim 7, it is characterized in that,

    The determination module (502) is further configured to determine the output of the production process and the logistics demand of the production process based on the electric energy consumption data in the production process;

    in

    

    Wherein the output of the production process is Q _k (t); the logistics demand of the production process is Cargo _k (t); Cap is the transportation capacity of each vehicle; P _k (t) is the Electric energy consumption data; Q _krated is the rated production capacity of the production process; P _krated is the rated electric power of the generation process; t is a time parameter.
11. The prediction device (500) of factory operation according to claim 7, it is characterized in that,

    The determination module (502) is configured to use the electric energy consumption data in the production process, the steam consumption in the production process and model input data together as training data to train the neural network model to obtain the electric energy consumption prediction A model; wherein said model input data includes at least one of the following: time stamp; weather; temperature; sewage production; sewage test results; holidays.
12. The prediction device (500) of factory operation according to claim 11, is characterized in that,

    The prediction module (503) is configured to input the model input data in the future production process into the electric energy consumption prediction model, so as to predict the electric energy consumption data in the future production process and the steam consumption data in the future production process; At least one of the following is determined based on the electrical energy consumption data in the future production process: predicted output of the future production process; predicted logistics requirements of the future production process; predicted storage requirements of the future production process.
13. A plant operation forecasting device (600), characterized by comprising a processor (601) and a memory (602);

    An application program executable by the processor (601) is stored in the memory (602), which is used to make the processor (601) perform the operation of the factory according to any one of claims 1 to 6. Prediction Methods (100).
14. A computer-readable storage medium, characterized in that computer-readable instructions are stored therein, and the computer-readable instructions are used to execute the method (100) for predicting plant operation according to any one of claims 1 to 6.

Images