CN115952853B - Method and device for constructing ore pulp density detection model and ore pulp density detection system - Google Patents
Method and device for constructing ore pulp density detection model and ore pulp density detection system Download PDFInfo
- Publication number
- CN115952853B CN115952853B CN202310238125.7A CN202310238125A CN115952853B CN 115952853 B CN115952853 B CN 115952853B CN 202310238125 A CN202310238125 A CN 202310238125A CN 115952853 B CN115952853 B CN 115952853B
- Authority
- CN
- China
- Prior art keywords
- pulp density
- density detection
- pulp
- detection model
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 166
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000012549 training Methods 0.000 claims description 34
- 238000012360 testing method Methods 0.000 claims description 21
- 238000003556 assay Methods 0.000 claims description 9
- 238000010276 construction Methods 0.000 claims description 5
- 239000002002 slurry Substances 0.000 claims description 5
- 230000001133 acceleration Effects 0.000 claims description 4
- 238000004519 manufacturing process Methods 0.000 abstract description 8
- SZVJSHCCFOBDDC-UHFFFAOYSA-N iron(II,III) oxide Inorganic materials O=[Fe]O[Fe]O[Fe]=O SZVJSHCCFOBDDC-UHFFFAOYSA-N 0.000 abstract description 3
- 238000005457 optimization Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 12
- 230000006870 function Effects 0.000 description 10
- 238000004590 computer program Methods 0.000 description 9
- 238000012545 processing Methods 0.000 description 6
- 238000003860 storage Methods 0.000 description 5
- 238000012544 monitoring process Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 239000012530 fluid Substances 0.000 description 3
- 230000036961 partial effect Effects 0.000 description 3
- 238000012937 correction Methods 0.000 description 2
- 229910052500 inorganic mineral Inorganic materials 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 230000000670 limiting effect Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 239000011707 mineral Substances 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000002829 reductive effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Paper (AREA)
Abstract
A method and a device for constructing an ore pulp density detection model and an ore pulp density detection system are provided, wherein the method for constructing the ore pulp density detection model comprises the following steps: acquiring historical data for constructing a pulp density detection model; according to historical data, respectively identifying a linear part and a nonlinear part of a pre-constructed initialized ore pulp density detection model to obtain a linear part model and a nonlinear part model; and combining the linear part model and the nonlinear part model to obtain a final ore pulp density detection model. By the method and the device for constructing the ore pulp density detection model and the ore pulp density detection system, the ore pulp density can be accurately detected in real time, the detection precision can meet the requirement of a production process on the ore pulp density detection precision, the detection cost is low, and conditions are created for realizing closed-loop optimization control of the magnetite ore pulp density.
Description
Technical Field
The invention relates to the technical field of detection of operation indexes of mineral processing industrial processes, in particular to a method and a device for constructing a pulp density detection model and a pulp density detection system.
Background
In the mineral separation process, the content of solid matters in ore pulp needs to be known in time so as to guide production. In the ore grinding process, the overflow granularity of the cyclone is a key technological parameter, and the ore feeding density directly influences the overflow granularity of the cyclone. The viscosity of ore pulp increases along with the increase of the density of ore pulp, and when the ore feeding density of the cyclone increases, the overflow granularity becomes coarse due to the increase of the viscosity, and the classification efficiency is affected. While the ore feeding density is small, the classification efficiency is high, but the ore pulp treatment capacity is reduced, and the cost consumption is increased. Thus, pulp density detection and control is essential in the separation process.
At present, two main methods exist for pulp density detection: manual assays and densitometer detection methods. The manual assay method has accurate results, but consumes a large amount of manpower and time resources, and the detected results have hysteresis and do not meet the requirements of real-time monitoring and controlling of the pulp density in the actual production process; densitometer detection, while faster than manual inspection, can be an unsatisfactory measurement accuracy in harsh industrial environments and complex operating conditions, and in addition, the densitometer is expensive and requires periodic maintenance, which also requires significant time and effort by personnel.
Therefore, the method has the unavoidable and neglectable defects of both the manual assay method and the densimeter detection method, and can not simultaneously meet the requirements of the ore pulp density detection accuracy and the real-time performance in the ore dressing process and the requirements of the industrial production process cost control.
Disclosure of Invention
In view of the above, the invention provides a method and a device for constructing an ore pulp density detection model and an ore pulp density detection system, and aims to solve the problem that the existing ore pulp density detection technology cannot meet the requirements of detection accuracy, real-time performance and cost control at the same time.
In a first aspect, the present invention provides a method for constructing a pulp density detection model, including: acquiring historical data for constructing a pulp density detection model; according to the historical data, respectively identifying a linear part and a nonlinear part of a pre-constructed initialized ore pulp density detection model to obtain a linear part model and a nonlinear part model; and combining the linear part model and the nonlinear part model to obtain a final ore pulp density detection model.
Further, the historical data for constructing the pulp density detection model comprises: pulp pump current, pulp pump frequency, pulp pressure and pulp density at each collection time.
Further, the ore pulp density detection model is constructed in advance by adopting the following mode:
ρ(t)=ρ 0 (t)+ Δρ(t)；
the linear part of the pulp density detection model is as follows:
ρ 0 (t)=k 1 P H (t)+k 2 P L (t)；
the nonlinear part of the pulp density detection model is as follows:
Δρ(t)=h(P H (t),P L (t),f(t),I(t))；
wherein ,P H (t) and P L (t) Respectively istThe high side absolute pressure and the low side absolute pressure at the moment,f(t) and I(t) Respectively istThe frequency and current of the slurry pump at the moment,h(·) As an unknown non-linear term,k 1 、k 2 is a linear model parameter.
Further, according to the historical data, identifying the linear part of the initialized pulp density detection model constructed in advance to obtain a linear part model, which comprises the following steps: and identifying the linear part of the initialized pulp density detection model constructed in advance by adopting a least square method according to the pulp pressure and the pulp density at each acquisition time in the historical data to obtain a linear part model.
Further, according to the pulp pressure and pulp density at each acquisition time in the historical data, a least square method is adopted to identify the linear part of the initialized pulp density detection model constructed in advance, and a linear part model is obtained, and the method comprises the following steps: training sample data are selected from the historical data, and the linear part of the initialized ore pulp density detection model constructed in advance is identified to obtain a linear part model as follows:
wherein the training data set is expressed as
；
tin order to acquire the time of day,t=k-1,k-2,…,kn, N is the number of training samples,kis the current time.
Further, according to the historical data, identifying a nonlinear part of a pre-constructed initialized pulp density detection model to obtain a nonlinear part model, wherein the method comprises the following steps: and identifying the nonlinear part of the initialized pulp density detection model constructed in advance by adopting a regularized random configuration network according to the pulp pump current, the pulp pump frequency, the pulp pressure and the pulp density at each acquisition time in the historical data to obtain a nonlinear part model.
Further, according to the pulp pump current, the pulp pump frequency, the pulp pressure and the pulp density at each acquisition time in the historical data, a regularized random configuration network is adopted to identify a nonlinear part of a pre-constructed initialized pulp density detection model, so as to obtain a nonlinear part model, which comprises the following steps: training sample data are selected from the historical data, and nonlinear parts of a pre-constructed initialized ore pulp density detection model are identified to obtain a nonlinear part model as follows:
wherein the training data set is
；
tin order to acquire the time of day,t=k-1,k-2,…,kn, N is the number of training samples,kis the current time.
In a second aspect, the present invention also provides a device for constructing a pulp density detection model, including: the acquisition unit is used for acquiring historical data for constructing the ore pulp density detection model; the identification unit is used for respectively identifying the linear part and the nonlinear part of the initialized ore pulp density detection model constructed in advance according to the historical data to obtain a linear part model and a nonlinear part model; and the merging unit is used for merging the linear part model and the nonlinear part model to obtain a final ore pulp density detection model.
In a third aspect, the present invention also provides a pulp density detection system, comprising: the terminal side control system is used for collecting field data at the current moment and sending the field data at the current moment to the side server; the side server is used for receiving the field data at the current moment and the pulp density detection model sent by the cloud side server, obtaining a pulp density predicted value at the current moment by adopting the pulp density detection model at the current moment based on the field data at the current moment, outputting the field data at the current moment, the pulp density predicted value at the current moment and the pulp density manual test value at the last moment, receiving the manual test value input by a user at the current moment, sending the field data at the current moment and the manual test value to the cloud side server, and storing the field data at the current moment, the pulp density predicted value and the manual test value into the database for inquiry; the cloud side server is used for acquiring historical data in the database, adopting the pulp density detection models obtained by the method for constructing the pulp density detection models according to the historical data, sending the pulp density detection models to the side server, receiving the field data and the manual test value at the current moment, and updating and comparing the pulp density detection models with accuracy based on the field data and the manual test value at the current moment to obtain the optimal model in real time.
According to the method and the device for constructing the ore pulp density detection model and the ore pulp density detection system, the linear part and the nonlinear part of the initialized ore pulp density detection model constructed in advance are respectively identified according to historical data, the linear part model and the nonlinear part model obtained through identification are combined to obtain the final ore pulp density detection model, and the ore pulp density detection model obtained through the method and the device for constructing the ore pulp density detection model and the ore pulp density detection system can realize accurate detection of the ore pulp density in real time, so that the detection precision can meet the requirement of a production process on the ore pulp density detection precision, the detection cost is low, and conditions are created for realizing closed-loop optimization control of the magnetite ore pulp density.
Drawings
FIG. 1 illustrates an exemplary flow chart of a method of pulp density detection model construction in accordance with an embodiment of the invention;
FIG. 2 shows a schematic algorithm of a pulp density detection model according to an embodiment of the invention;
FIG. 3 shows a comparison of accuracy of test results obtained using a pulp density test model in accordance with an embodiment of the present invention;
FIG. 4 is a schematic diagram showing the structure of an apparatus for constructing a pulp density detection model according to an embodiment of the present invention;
FIG. 5 shows a schematic diagram of a pulp density detection system according to an embodiment of the invention;
FIG. 6 shows a schematic diagram of a specific architecture of a pulp density detection system according to an embodiment of the invention;
fig. 7 shows an interface schematic of a pulp density detection system according to an embodiment of the invention.
Detailed Description
The exemplary embodiments of the present invention will now be described with reference to the accompanying drawings, however, the present invention may be embodied in many different forms and is not limited to the examples described herein, which are provided to fully and completely disclose the present invention and fully convey the scope of the invention to those skilled in the art. The terminology used in the exemplary embodiments illustrated in the accompanying drawings is not intended to be limiting of the invention. In the drawings, like elements/components are referred to by like reference numerals.
Unless otherwise indicated, terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art. In addition, it will be understood that terms defined in commonly used dictionaries should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense.
Fig. 1 shows an exemplary flow chart of a method of constructing a pulp density detection model according to an embodiment of the invention.
As shown in fig. 1, the method for constructing the pulp density detection model comprises the following steps:
step S101: historical data for constructing a pulp density detection model is obtained.
Further, the historical data for constructing the pulp density detection model includes: pulp pump current, pulp pump frequency, pulp pressure and pulp density at each collection time.
If the frequency of the data collected by the pressure sensor and the frequency converter in the pulp flowing process iszSecond, orderzIn a unit, record the current time askCollectingt=k-1,k-2 ,…,k-N（NSample number), the data of pulp pump current, pulp pump frequency, pulp pressure and pulp density at the moment are used as training sets. Wherein, the pulp pressure includes high pressure side absolute pressure and low pressure side absolute pressure.
Step S102: and respectively identifying the linear part and the nonlinear part of the initialized pulp density detection model constructed in advance according to the historical data to obtain a linear part model and a nonlinear part model.
Further, the ore pulp density detection model is constructed in advance by adopting the following mode:
ρ(t)=ρ 0 (t)+ Δρ(t)；
the linear part of the pulp density detection model is as follows:
ρ 0 (t)=k 1 P H (t)+k 2 P L (t)；
the nonlinear part of the pulp density detection model is as follows:
Δρ(t)=h(P H (t),P L (t),f(t),I(t))；
wherein ,P H (t) and P L (t) Respectively istThe high side absolute pressure and the low side absolute pressure at the moment,f(t) and I(t) Respectively istThe frequency and current of the slurry pump at the moment,h(·) As an unknown non-linear term,k 1 、k 2 is a parameter of the linear model.
Under ideal conditions without loss of resistance, the differential pressure is:
in the formula ：
for pulp density>
Acceleration of gravity, ++>
Is the liquid level difference.
Further, considering the along-path resistance loss of ore pulp in the flowing process, the calculation formula is as follows:
in the formula ：
for the resistance coefficient along the way>
Is the length of the tube>
For the diameter of the pipeline>
For average flow velocity of section>
Gravitational acceleration.
In an actual industrial process, the total pressure difference
It is possible to obtain:
further, according to Bernoulli's law, the total pressure of the fluid is unchanged. When the flow rate becomes large, the static pressure of the fluid becomes small, and the pulp pressure meter can only measure the static pressure, and the fluid flow process is simulated to study the relationship between the dynamic pressure difference and the flow rate. In addition, the pressure meter can generate systematic errors and random errors when measuring the pressure, so that the measured absolute pressure of the high pressure side is consideredP H And low side absolute pressureP L When the correction is performed, the differential pressure measured at time t is measured
The rewriting is as follows:
in the formula ：
is unknown constant (I)>
Representing an unknown nonlinear error in measuring pressure.
Average flow rate
Current to pulp pumpISum frequencyfThe unknown nonlinear expression of (2) is as follows:
further, the following formulas (4) and (5) are taken to be (3):
wherein ,
；
in the formula ：h(·) The representation comprises unknown nonlinear terms such as instrument measurement errors, unknown disturbance in the pulp flowing process and the like.
Further, according to the historical data, identifying the linear part of the initialized pulp density detection model constructed in advance to obtain a linear part model, which comprises the following steps:
and identifying the linear part of the initialized pulp density detection model constructed in advance by adopting a least square method according to the pulp pressure and the pulp density at each acquisition time in the historical data to obtain a linear part model.
Further, according to the pulp pressure and pulp density at each acquisition time in the historical data, a least square method is adopted to identify the linear part of the initialized pulp density detection model constructed in advance, and a linear part model is obtained, and the method comprises the following steps:
training sample data are selected from the historical data, and the linear part of the initialized ore pulp density detection model constructed in advance is identified to obtain a linear part model as follows:
wherein the training data set is expressed as
；
tin order to acquire the time of day,t=k-1,k-2,…,kn, N is the number of training samples,kis the current time.
Specifically, the algorithm structure is as shown in fig. 2, and the acquisition is performedt=k-1,k-2,…,k-data at time N as training samples, N being the number of training samples.
The training dataset is represented as
；
Is provided with
，
Then there are:
is provided with
The loss function of the model estimate is: />
When the derivative of the loss function is 0, an extremum is taken, namely:
after that, get
The analytical solution of (2) is:
the linear partial model is then expressed as:
further, according to the historical data, identifying the nonlinear part of the initialized pulp density detection model constructed in advance to obtain a nonlinear part model, which comprises the following steps:
and identifying the nonlinear part of the initialized pulp density detection model built in advance by adopting a regularized random configuration network according to the pulp pump current, the pulp pump frequency, the pulp pressure and the pulp density at each acquisition time in the historical data to obtain a nonlinear part model.
Further, according to the pulp pump current, the pulp pump frequency, the pulp pressure and the pulp density at each acquisition time in the historical data, a regularized random configuration network is adopted to identify a nonlinear part of a pre-constructed initialized pulp density detection model, so as to obtain a nonlinear part model, which comprises the following steps:
training sample data are selected from the historical data, and nonlinear parts of the initialized ore pulp density detection model constructed in advance are identified to obtain a nonlinear part model as follows:
wherein the training data set is
；
tin order to acquire the time of day,t=k-1,k-2,…,kn, N is the number of training samples,kis the current time.
Specifically, the algorithm structure is as shown in fig. 2, and the acquisition is performedt=k-1,k-2,…,k-data at time N as training samples, N being the number of training samples.
Training data set is
；
tin order to acquire the time of day,t=k-1,k-2,…,kn, N is the number of training samples,kis the current time.
First, an objective function is given
Assuming that the nodes of the current hidden layer are L-1, the neural network output result can be represented by equation (12): />
in the formula ：
representing hidden layer nodesjOutput weights of (2); />
(/>
) Representing an activation function; />
and />
Respectively represent hidden layerjThe input weights and biases of the hidden nodes.
Thereafter, computing a network output residual:
then, the maximum hidden layer node number is
The desired accuracy is +.>
. If->
Then add->
Hidden layer nodes are obtained from formulas (14) and (15), respectively ++>
Output of +.>
And supervision mechanism->
Formulas (14), (15) are as follows:
in the formula ：
given->
Let->
，/>
Is a regular term coefficient. Random arrangement->
Every time the configuration will randomly select the input weight of the hidden layer within a certain range +.>
And bias->
And calculate +.>
. If->
Will->
Store, if all->
Are not in conformity with the condition, then a larger one is selectedrValue reconfiguration. After the random configuration is finished, the maximum +.>
Corresponding->
As->
Input weights and biases for the individual nodes.
Thereafter, the output weight is calculated by the formula (16)
：
in the formula ,
，Iis an identity matrix.
Thereafter, get at
The time of day, the nonlinear partial model is expressed as:
step S103: and combining the linear part model and the nonlinear part model to obtain a final ore pulp density detection model.
Bringing the resulting linear and nonlinear partial models into equation (6), i.eρ(t)=ρ 0 (t)+ Δρ(t) And obtaining a final ore pulp density detection model.
Further, the method for constructing the ore pulp density detection model further comprises the following steps:
the historical data is updated in real-time for model updating.
Further, updating the historical data in real time for model updating includes:
acquiring field data at the current moment and a manual assay value input by a user;
and updating the existing pulp density detection model according to the field data at the current moment and the manual test value input by the user, and comparing the updated pulp density detection model with the pulp density before updating to obtain an optimal model to be used as the pulp density detection model at the next moment.
And acquiring data acquired in real time on site, such as information real-time monitoring data of pulp pump current, pulp pump frequency, pulp pressure, pulp density and the like, acquiring a manual assay value input by a user at an interactive interface, and training and updating an existing model according to the updated historical data.
Specifically, the historical dataset has 197 sets of data together, according to the training set: test set = 6:4, based on the historical data, the pulp density detection model obtained by the method for constructing the pulp density detection model provided by the embodiment is adopted to detect the pulp density, the detection result is shown in fig. 3, and the degree of coincidence between the model predicted value and the artificial assay value can be seen.
According to the embodiment, the linear part and the nonlinear part of the initialized ore pulp density detection model which are built in advance are respectively identified according to the historical data, and the linear part model and the nonlinear part model which are obtained through identification are combined to obtain the final ore pulp density detection model.
Fig. 4 shows a schematic structural diagram of an apparatus for constructing a pulp density detection model according to an embodiment of the invention.
As shown in fig. 4, the pulp density detection model construction apparatus includes:
an acquisition unit 401 for acquiring historical data for constructing a pulp density detection model;
an identification unit 402, configured to identify a linear portion and a nonlinear portion of the initialized pulp density detection model constructed in advance according to the historical data, so as to obtain a linear portion model and a nonlinear portion model;
and the merging unit 403 is configured to merge the linear part model and the nonlinear part model to obtain a final pulp density detection model.
Further, the historical data for constructing the pulp density detection model comprises: pulp pump current, pulp pump frequency, pulp pressure and pulp density at each collection time.
Further, the ore pulp density detection model is constructed in advance by adopting the following mode:
ρ(t)=ρ 0 (t)+ Δρ(t)；
the linear part of the pulp density detection model is as follows:
ρ 0 (t)=k 1 P H (t)+k 2 P L (t)；
the nonlinear part of the pulp density detection model is as follows:
Δρ(t)=h(P H (t),P L (t),f(t),I(t))；
wherein ,P H (t) and P L (t) Respectively istThe high side absolute pressure and the low side absolute pressure at the moment,f(t) and I(t) Respectively istThe frequency and current of the slurry pump at the moment,h(·) As an unknown non-linear term,k 1 、k 2 is a parameter of the linear model.
Further, according to the historical data, identifying the linear part of the initialized pulp density detection model constructed in advance to obtain a linear part model, which comprises the following steps:
and identifying the linear part of the initialized pulp density detection model constructed in advance by adopting a least square method according to the pulp pressure and the pulp density at each acquisition time in the historical data to obtain a linear part model.
Further, according to the pulp pressure and pulp density at each acquisition time in the historical data, a least square method is adopted to identify the linear part of the initialized pulp density detection model constructed in advance, and a linear part model is obtained, and the method comprises the following steps:
training sample data are selected from the historical data, and the linear part of the initialized ore pulp density detection model constructed in advance is identified to obtain a linear part model as follows:
wherein the training data set is expressed as
；
tin order to acquire the time of day,t=k-1,k-2,…,kn, N is the number of training samples,kis the current time.
Further, according to the historical data, identifying the nonlinear part of the initialized pulp density detection model constructed in advance to obtain a nonlinear part model, which comprises the following steps:
and identifying the nonlinear part of the initialized pulp density detection model built in advance by adopting a regularized random configuration network according to the pulp pump current, the pulp pump frequency, the pulp pressure and the pulp density at each acquisition time in the historical data to obtain a nonlinear part model.
Further, according to the pulp pump current, the pulp pump frequency, the pulp pressure and the pulp density at each acquisition time in the historical data, a regularized random configuration network is adopted to identify a nonlinear part of a pre-constructed initialized pulp density detection model, so as to obtain a nonlinear part model, which comprises the following steps:
training sample data are selected from the historical data, and nonlinear parts of the initialized ore pulp density detection model constructed in advance are identified to obtain a nonlinear part model as follows:
wherein the training data set is
；
tin order to acquire the time of day,t=k-1,k-2,…,kn, N is the number of training samples,kis the current time.
It should be noted that, when the apparatus provided in the foregoing embodiment performs the functions thereof, only the division of the foregoing functional modules is used as an example, in practical application, the foregoing functional allocation may be performed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules, so as to perform all or part of the functions described above. In addition, the apparatus and the method embodiments provided in the foregoing embodiments belong to the same concept, and specific implementation processes of the apparatus and the method embodiments are detailed in the method embodiments and are not repeated herein.
According to the embodiment, the linear part and the nonlinear part of the initialized ore pulp density detection model which are built in advance are respectively identified according to the historical data, and the linear part model and the nonlinear part model which are obtained through identification are combined to obtain the final ore pulp density detection model.
Fig. 5 shows a schematic diagram of the structure of the pulp density detection system according to an embodiment of the invention.
As shown in fig. 5, the pulp density detection system includes:
the end-side control system 501 is configured to collect field data at a current time and send the field data at the current time to an edge-side server;
the side server 502 is configured to receive field data at a current time and a pulp density detection model sent by the cloud side server, obtain a pulp density prediction value at the current time by using the pulp density detection model at the current time based on the field data at the current time, output the field data at the current time, the pulp density prediction value at the current time and a pulp density manual test value at a previous time, receive a manual test value input by a user at the current time, send the field data at the current time and the manual test value to the cloud side server, and store the field data at the current time, the pulp density prediction value and the manual test value in a database for query;
the cloud side server 503 is configured to obtain historical data in the database, according to the historical data, send the pulp density detection model to the side server by using the pulp density detection model construction method provided by the above embodiments, receive field data and a manual test value at the current moment, and update and compare accuracy of the pulp density detection model based on the field data and the manual test value at the current moment to obtain an optimal model in real time.
Fig. 6 shows a schematic diagram of a specific architecture of a pulp density detection system according to an embodiment of the invention. As shown in fig. 6, the end side is connected with a detection instrument and a frequency converter through an analog input module by a PLC controller, and the pressure instrument, the pulp pump current and the frequency data of the industrial site are collected. And executing data processing and an online intelligent detection model by using an edge control system at the side, monitoring information such as pulp pump current, pulp pump frequency, pulp pressure, pulp density and the like by an operator through a monitoring interface, and submitting a manual assay value at an interactive interface for model updating. And running pulp density detection software on the cloud side by using a virtual machine in the informationized cloud platform, connecting the software with a database, reading and storing data, acquiring processed data from the side, updating the model offline, evaluating the model, and updating the model with higher precision to the side for online detection. Fig. 7 shows an interface schematic of a pulp density detection system according to an embodiment of the invention. As shown in fig. 7, the interface outputs the field data at the current time, the pulp density predicted value at the current time, and the pulp density manual test value at the previous time, and may output the model correction times.
According to the embodiment, through adopting the 'end edge cloud' integrated cooperative pulp density detection system, the accurate detection of the pulp density can be realized in real time, the detection precision can meet the requirements of a production process on the pulp density detection precision, the detection cost is low, and conditions are created for realizing closed-loop optimization control of the magnetite pulp density.
The embodiment of the invention also provides a computer readable storage medium, and a computer program is stored on the computer readable storage medium, and when the computer program is executed by a processor, the method for constructing the ore pulp density detection model provided by each embodiment is realized.
The embodiment of the invention also provides electronic equipment, which comprises: a processor; a memory for storing processor-executable instructions; the processor is used for reading the executable instructions from the memory and executing the instructions to realize the method for constructing the ore pulp density detection model provided by each embodiment.
The invention has been described with reference to a few embodiments. However, as is well known to those skilled in the art, other embodiments than the above disclosed invention are equally possible within the scope of the invention, as defined by the appended patent claims.
Generally, all terms used in the claims are to be interpreted according to their ordinary meaning in the technical field, unless explicitly defined otherwise therein. All references to "a/an/the [ means, component, etc. ]" are to be interpreted openly as referring to at least one instance of said means, component, etc., unless explicitly stated otherwise. The steps of any method disclosed herein do not have to be performed in the exact order disclosed, unless explicitly stated.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Finally, it should be noted that: the above embodiments are only for illustrating the technical aspects of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the above embodiments, it should be understood by those of ordinary skill in the art that: modifications and equivalents may be made to the specific embodiments of the invention without departing from the spirit and scope of the invention, which is intended to be covered by the claims.
Claims (5)
1. The method for constructing the pulp density detection model is characterized by comprising the following steps of:
acquiring historical data for constructing a pulp density detection model; the historical data for constructing the ore pulp density detection model comprises the following steps: pulp pump current, pulp pump frequency, pulp density, high-pressure side absolute pressure and low-pressure side absolute pressure at each acquisition time;
identifying a linear part of a pre-built initialized pulp density detection model by adopting a least square method according to the pulp pressure and the pulp density at each acquisition time in the historical data to obtain a linear part model, and identifying a nonlinear part of the pre-built initialized pulp density detection model by adopting a regularized random configuration network according to the pulp pump current, the pulp pump frequency, the pulp pressure and the pulp density at each acquisition time in the historical data to obtain a nonlinear part model;
combining the linear part model and the nonlinear part model to obtain a final ore pulp density detection model;
the ore pulp density detection model is constructed in advance by adopting the following mode:
ρ(t)= ρ 0 (t)+ Δρ(t)；
the linear part of the pulp density detection model is as follows:
ρ 0 (t)= k 1 P H (t)+ k 2 P L (t)；
the nonlinear part of the pulp density detection model is as follows:
Δρ (t)= h(P H (t), P L (t), f(t), I(t))；
wherein ,P H (t) and P L (t) Respectively istThe high side absolute pressure and the low side absolute pressure at the moment,f(t) and I(t) Respectively istThe frequency and current of the slurry pump at the moment,h (·) As an unknown non-linear term,k 1 、k 2 parameters of a linear model;
wherein ,ρ(t)= ρ 0 (t)+ Δρ(t) The method is characterized by comprising the following steps:
in the formula ,tdifferential pressure delta measured at time of dayPThe following are provided:
in the formula ：
is unknown constant (I)>
Representing unknown nonlinear errors in measuring pressureDifference;
tthe ore pulp generates the resistance loss along the way in the flowing process at the momenth f The following are provided:
in the formula ：
for the resistance coefficient along the way>
Is the length of the tube>
For the diameter of the pipeline>
Gravitational acceleration; />
Is thattThe average flow velocity of the section at the moment is as follows:
2. the method according to claim 1, wherein the step of identifying the linear part of the initialized pulp density detection model constructed in advance by using a least square method according to the pulp pressure and the pulp density at each acquisition time in the history data to obtain a linear part model comprises the steps of:
training sample data are selected from the historical data, and the linear part of the initialized ore pulp density detection model constructed in advance is identified to obtain a linear part model as follows:
wherein the training data set is expressed as
；
tin order to acquire the time of day,t=k-1,k-2,…,kn, N is the number of training samples,kis the current time.
3. The method for constructing a pulp density detection model according to claim 1, wherein identifying the nonlinear part of the initialized pulp density detection model constructed in advance by using a regularized random configuration network according to the pulp pump current, the pulp pump frequency, the pulp pressure and the pulp density at each acquisition time in the historical data to obtain the nonlinear part model comprises the following steps:
training sample data are selected from the historical data, and nonlinear parts of a pre-constructed initialized ore pulp density detection model are identified to obtain a nonlinear part model as follows:
wherein the training data set is
；
tin order to acquire the time of day,t=k-1,k-2,…,kn, N is the number of training samples,kis the current time.
4. The utility model provides a pulp density detects model construction device which characterized in that includes:
the acquisition unit is used for acquiring historical data for constructing the ore pulp density detection model; the historical data for constructing the ore pulp density detection model comprises the following steps: pulp pump current, pulp pump frequency, pulp density, high-pressure side absolute pressure and low-pressure side absolute pressure at each acquisition time;
the identification unit is used for identifying the linear part of the initialized pulp density detection model which is built in advance by adopting a least square method according to the pulp pressure and the pulp density at each acquisition time in the historical data to obtain a linear part model, and identifying the nonlinear part of the initialized pulp density detection model which is built in advance by adopting a regularized random configuration network according to the pulp pump current, the pulp pump frequency, the pulp pressure and the pulp density at each acquisition time in the historical data to obtain a nonlinear part model;
the merging unit is used for merging the linear part model and the nonlinear part model to obtain a final ore pulp density detection model;
the ore pulp density detection model is constructed in advance by adopting the following mode:
ρ(t)= ρ 0 (t)+ Δρ(t)；
the linear part of the pulp density detection model is as follows:
ρ 0 (t)= k 1 P H (t)+ k 2 P L (t)；
the nonlinear part of the pulp density detection model is as follows:
Δρ (t)= h(P H (t), P L (t), f(t), I(t))；
wherein ,P H (t) and P L (t) Respectively istThe high side absolute pressure and the low side absolute pressure at the moment,f(t) and I(t) Respectively istThe frequency and current of the slurry pump at the moment,h (·) As an unknown non-linear term,k 1 、k 2 parameters of a linear model;
wherein ,ρ(t)= ρ 0 (t)+ Δρ(t) The method is characterized by comprising the following steps:
in the formula ,tdifferential pressure delta measured at time of dayPThe following are provided:
in the formula ：
is unknown constant (I)>
Representing an unknown nonlinear error in measuring pressure;
tthe ore pulp generates the resistance loss along the way in the flowing process at the momenth f The following are provided:
in the formula ：
for the resistance coefficient along the way>
Is the length of the tube>
For the diameter of the pipeline>
Gravitational acceleration; />
Is thattThe average flow velocity of the section at the moment is as follows:
5. a pulp density detection system, comprising:
the terminal side control system is used for collecting field data at the current moment and sending the field data at the current moment to the side server;
the side server is used for receiving the field data at the current moment and the pulp density detection model sent by the cloud side server, obtaining a pulp density predicted value at the current moment by adopting the pulp density detection model at the current moment based on the field data at the current moment, outputting the field data at the current moment, the pulp density predicted value at the current moment and the pulp density manual test value at the last moment, receiving the manual test value input by a user at the current moment, sending the field data at the current moment and the manual test value to the cloud side server, and storing the field data at the current moment, the pulp density predicted value and the manual test value into the database for inquiry;
the cloud side server is used for acquiring historical data in the database, adopting the pulp density detection model obtained by the pulp density detection model construction method according to any one of claims 1-3 according to the historical data, sending the pulp density detection model to the side server, receiving the field data and the manual assay value at the current moment, and updating and comparing the pulp density detection model with accuracy based on the field data and the manual assay value at the current moment to obtain an optimal model in real time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310238125.7A CN115952853B (en) | 2023-03-14 | 2023-03-14 | Method and device for constructing ore pulp density detection model and ore pulp density detection system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310238125.7A CN115952853B (en) | 2023-03-14 | 2023-03-14 | Method and device for constructing ore pulp density detection model and ore pulp density detection system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115952853A CN115952853A (en) | 2023-04-11 |
| CN115952853B true CN115952853B (en) | 2023-05-26 |
Family
ID=85903337
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310238125.7A Active CN115952853B (en) | 2023-03-14 | 2023-03-14 | Method and device for constructing ore pulp density detection model and ore pulp density detection system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115952853B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102735579A (en) * | 2011-03-24 | 2012-10-17 | 阿自倍尔株式会社 | Density measuring system and density measuring method |
| CN110411891A (en) * | 2019-09-04 | 2019-11-05 | 上海乐研电气有限公司 | Realize the non-maintaining on-site detecting device of gas density relay, system and method |
| CN115083123A (en) * | 2022-05-17 | 2022-09-20 | 中国矿业大学 | Mine coal spontaneous combustion intelligent grading early warning method taking measured data as drive |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109446669B (en) * | 2018-11-01 | 2022-10-28 | 东北大学 | Soft measurement method for ore pulp concentration |
| CN113607601B (en) * | 2021-06-18 | 2022-10-28 | 东北大学 | Intelligent detection method for ore pulp concentration based on combination of identification model and deep learning |
| CN114626304B (en) * | 2022-03-21 | 2024-02-27 | 齐鲁工业大学 | Online prediction soft measurement modeling method for ore pulp copper grade |
| CN115598979A (en) * | 2022-10-19 | 2023-01-13 | 三一重机有限公司（Cn） | Method and device for identifying model parameters of hydraulic system and hydraulic engineering machinery |
-
2023
- 2023-03-14 CN CN202310238125.7A patent/CN115952853B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102735579A (en) * | 2011-03-24 | 2012-10-17 | 阿自倍尔株式会社 | Density measuring system and density measuring method |
| CN110411891A (en) * | 2019-09-04 | 2019-11-05 | 上海乐研电气有限公司 | Realize the non-maintaining on-site detecting device of gas density relay, system and method |
| CN115083123A (en) * | 2022-05-17 | 2022-09-20 | 中国矿业大学 | Mine coal spontaneous combustion intelligent grading early warning method taking measured data as drive |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115952853A (en) | 2023-04-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116150897A (en) | Machine tool spindle performance evaluation method and system based on digital twin | |
| CN114970688A (en) | Landslide monitoring data preprocessing method based on LSTMAD algorithm and Hermite interpolation method | |
| CN115238394B (en) | Multi-source uncertainty hybrid reliability digital twin modeling method for composite material structure | |
| CN114265001B (en) | Smart electric meter metering error evaluation method | |
| EP4064138A1 (en) | Apparatus, method, and program for evaluating the deterioration of an object | |
| CN117556366B (en) | Data abnormality detection system and method based on data screening | |
| CN113570165B (en) | Intelligent prediction method for permeability of coal reservoir based on particle swarm optimization | |
| CN115952853B (en) | Method and device for constructing ore pulp density detection model and ore pulp density detection system | |
| CN101793620B (en) | Health monitoring method of cable system based on cable force monitoring during support settlement | |
| CN113607601A (en) | Intelligent detection method for ore pulp concentration based on combination of identification model and deep learning | |
| CN109101759A (en) | A kind of parameter identification method based on forward and reverse response phase method | |
| CN115839344A (en) | Wear monitoring method, device, equipment and storage medium for slurry pump | |
| CN111062118B (en) | Multilayer soft measurement modeling system and method based on neural network prediction layering | |
| CN110705186B (en) | Real-time online instrument checksum diagnosis method through RBF particle swarm optimization algorithm | |
| CN115879355A (en) | Temperature compensation method of piezoelectric sensor | |
| CN113128053A (en) | Nonlinear system parameter identification method, device, equipment and medium | |
| CN110750756B (en) | Real-time on-line instrument checksum diagnosis method through optimal support vector machine algorithm | |
| CN114048925A (en) | Power grid comprehensive operation early warning method and device and terminal equipment | |
| CN113532614A (en) | Method, processor and weighing system for predicting sensor data | |
| CN101789054A (en) | Health monitoring method of cable system based on space coordinate monitoring during support settlement | |
| Zimoch et al. | Process of hydraulic models calibration | |
| CN115511341B (en) | Method and device for evaluating time-varying failure probability of reservoir bank slope | |
| CN107607182A (en) | A kind of truck weighing system and Weighing method | |
| Al-Adly et al. | Physics-informed neural networks for structural health monitoring: a case study for Kirchhoff–Love plates | |
| CN115062273B (en) | Photoelectric sensor precision control method and system for industrial internet |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |