CN110454904B - Method and device for processing refrigerant content in dehumidifier - Google Patents
Method and device for processing refrigerant content in dehumidifier Download PDFInfo
- Publication number
- CN110454904B CN110454904B CN201810427239.5A CN201810427239A CN110454904B CN 110454904 B CN110454904 B CN 110454904B CN 201810427239 A CN201810427239 A CN 201810427239A CN 110454904 B CN110454904 B CN 110454904B
- Authority
- CN
- China
- Prior art keywords
- content
- dehumidifier
- refrigerant
- loss value
- preset
- 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
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/30—Control or safety arrangements for purposes related to the operation of the system, e.g. for safety or monitoring
Abstract
The invention discloses a method and a device for processing the content of a refrigerant in a dehumidifier. Wherein, the method comprises the following steps: acquiring the operating parameters of the dehumidifier; analyzing the operation parameters by using a classification model to determine the probability that the content of the refrigerant in the dehumidifier is various target contents; and determining the actual content of the refrigerant in the dehumidifier based on the probability of the various target contents and a preset loss value, wherein the preset loss value is used for representing the loss caused by judging the target content as the first content by the classification model, and the first content is any one of the various target contents. The invention solves the technical problem of poor reliability of the method for processing the content of the refrigerant in the dehumidifier in the prior art.
Description
Technical Field
The invention relates to the field of dehumidifier detection, in particular to a method and a device for processing the content of refrigerant in a dehumidifier.
Background
At present, a fluorine-deficient detection method of a dehumidifier is established based on expert experience on the basis of observing and researching the actual operation parameters of a machine. Due to the fact that the internal structure of the dehumidifier is complex, the coupling relation and the change rule among the state quantities in the operation process are difficult to master completely, and the expert experience has certain subjectivity and other reasons, the control rule is complex, and the generalization capability needs to be improved.
To improve this, the fluorine content of the dehumidifier can be predicted based on some classification algorithm to fit the relationship between the factors related to the fluorine content of the dehumidifier. In the existing machine learning technology, due to the interference of many random factors, the result of the classification algorithm still cannot reach 100% accuracy, that is, the probability of prediction error still exists when the fluorine content of the dehumidifier is predicted by the classification algorithm. Further, the loss after the prediction error is different for each fluorine content. For example, if the fluorine content of the dehumidifier is determined to be 60% and the fluorine is required to be reported to be lacking, then if 50% is reported to be 60% at the moment, even though the error is caused, the fluorine is also lacking; if the prediction is wrong from 70% to 60%, the problem of fluorine deficiency and false alarm is solved, and thus some losses are brought to enterprises; if the report rate is 70% from 60%, the problem of fluorine-lacking false report is also caused, and some bad experiences are brought to the user, namely, errors are separated by 10%, but the loss is different.
Aiming at the problem of poor reliability of a processing method of the refrigerant content in the dehumidifier in the prior art, no effective solution is provided at present.
Disclosure of Invention
The embodiment of the invention provides a method and a device for processing the content of a refrigerant in a dehumidifier, which at least solve the technical problem of poor reliability of a method for processing the content of the refrigerant in the dehumidifier in the prior art.
According to an aspect of an embodiment of the present invention, a method for processing a refrigerant content in a dehumidifier is provided, including: acquiring the operating parameters of the dehumidifier; analyzing the operation parameters by using a classification model to determine the probability that the content of the refrigerant in the dehumidifier is various target contents; and determining the actual content of the refrigerant in the dehumidifier based on the probability of the various target contents and a preset loss value, wherein the preset loss value is used for representing the loss caused by judging the target content as the first content by the classification model, and the first content is any one of the various target contents.
Further, determining the actual content of the refrigerant in the dehumidifier based on the probability of the various target contents and the preset loss value, wherein the determining comprises the following steps: obtaining a total loss value of the first content according to the probabilities of the various target contents and a preset loss value; obtaining a minimum total loss value of the total loss values of the plurality of first contents; and determining the first content corresponding to the minimum total loss value as the actual content.
Further, obtaining a total loss value of the first content according to the probabilities of the various target contents and the preset loss value, including: obtaining the product of the probability of various target contents and a preset loss value to obtain a plurality of products; and obtaining the sum of the products to obtain the total loss value of the first content.
Further, the total loss value Y (α) of the first content is obtained by the following formulai|X)：
Wherein i, j is 1,2,3, …, n, n is the number of the target contents, alphaiDenotes a first content, p (ω)j| X) is that the content X of the refrigerant in the dehumidifier is determined as omegajProbability, ω, corresponding to target contentjIs a target ofContent, beta (. alpha.)iJ) is a preset loss value, and X is the content of the refrigerant in the dehumidifier.
Further, the obtaining of the operation parameters of the dehumidifier includes: and after the dehumidifier normally operates for a first preset time, acquiring operation parameters every second preset time to obtain a plurality of operation parameters of the dehumidifier.
Further, determining the actual content of the refrigerant in the dehumidifier based on the probability of the various target contents and the preset loss value, wherein the determining comprises the following steps: determining a plurality of contents of the refrigerant in the dehumidifier based on the probabilities of the plurality of target contents and a preset loss value; obtaining the proportion of various target contents according to the contents; obtaining the maximum proportion of the proportions of the various target contents; and determining the target content corresponding to the maximum proportion as the actual content.
Further, before analyzing the operation parameters by using the classification model and determining the probability that the content of the refrigerant in the dehumidifier is various target contents, the method further comprises the following steps: establishing an initial classification model; acquiring multiple groups of training sample data, wherein each group of training sample data comprises: the target content and the actual content of the refrigerant are the operation parameters of the dehumidifier with the target content under different working conditions; and training the initial classification model through multiple groups of training sample data to obtain a classification model.
Further, after the initial classification model is trained through a plurality of sets of training sample data to obtain a classification model, the method further includes: acquiring multiple groups of test sample data, wherein each group of test sample data comprises: the target content and the actual content of the refrigerant are the operation parameters of the dehumidifier with the target content under different working conditions; testing the classification model through a plurality of groups of test sample data, and judging whether the output of the classification model meets a first preset condition; and if the output of the classification model does not meet the first preset condition, continuing training the classification model through multiple groups of training sample data until the output of the classification model meets the first preset condition.
Further, after determining that the output of the classification model satisfies the first preset condition, the method further includes: determining the actual content of the refrigerant in the dehumidifier based on the probability of various target contents output by the classification model and a preset loss value; judging whether the actual content meets a second preset condition or not; and if the actual content is determined not to meet the second preset condition, adjusting the preset loss value, and determining the actual content based on the probabilities of the various target contents and the adjusted preset loss value until the actual content meets the second preset condition.
Further, before determining that the actual content meets the second preset condition, the method further includes: and storing the classification model and the preset loss value to a server and a controller of the dehumidifier.
Further, before obtaining the operation parameters of the dehumidifier, the method further comprises: judging whether the dehumidifier has a networking function or not; if the dehumidifier has the networking function, the operation parameters are sent to the slave server, the actual content returned by the server is received, or the operation parameters are processed through a controller of the dehumidifier to obtain the actual content; and if the dehumidifier does not have the networking function, processing the operation parameters through a controller of the dehumidifier to obtain the actual content.
Further, after determining the actual content of the refrigerant in the dehumidifier based on the probability of the plurality of target contents and the preset loss value, the method further comprises: judging whether the actual content is lower than a preset value; and if the actual content is lower than the preset value, generating alarm prompt information or controlling the dehumidifier to execute preset action, wherein the alarm prompt information is used for prompting that the actual content is lower than the preset value.
According to another aspect of the embodiments of the present invention, there is also provided a device for processing the content of refrigerant in a dehumidifier, including: the acquisition module is used for acquiring the operating parameters of the dehumidifier; the first determination module is used for analyzing the operation parameters by using the classification model and determining the probability that the content of the refrigerant in the dehumidifier is various target contents; and the second determination module is used for determining the actual content of the refrigerant in the dehumidifier based on the probabilities of various target contents and a preset loss value, wherein the preset loss value is used for representing the loss caused by judging the target content as the first content by the classification model, and the first content is any one of the various target contents.
According to another aspect of the embodiments of the present invention, there is also provided a storage medium, where the storage medium includes a stored program, and when the program runs, the device where the storage medium is located is controlled to execute the method for processing the content of the media in the dehumidifier.
According to another aspect of the embodiments of the present invention, there is also provided a processor, where the processor is configured to execute a program, where the program executes the method for processing the content of the refrigerant in the dehumidifier.
In the embodiment of the invention, after the operation parameters of the dehumidifier are obtained, the operation parameters can be analyzed by using a classification model, the probability that the content of the refrigerant in the dehumidifier is various target contents is determined, and the actual content of the refrigerant in the dehumidifier is further determined based on the probability of the various target contents and the preset loss value, so that the prediction of the content of the refrigerant in the dehumidifier is realized. Because the minimum risk decision function is added after the probabilities of various target contents are output through the classification model, the actual content of the refrigerant in the dehumidifier is determined through the probabilities of the various target contents and the preset loss value, the loss caused by prediction errors of the classification algorithm is reduced, the processing accuracy and reliability of the refrigerant content in the dehumidifier are improved, and the technical problem of poor reliability of the processing method of the refrigerant content in the dehumidifier in the prior art is solved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:
FIG. 1 is a flow chart of a method for processing refrigerant content in a dehumidifier according to an embodiment of the present invention;
FIG. 2 is a schematic structural diagram of an alternative method for processing refrigerant content in a dehumidifier, according to an embodiment of the present invention;
FIG. 3 is a flow chart of an alternative classification model training and preset penalty value adjustment according to an embodiment of the present invention;
FIG. 4 is a flow chart of an alternative method of processing refrigerant content in a dehumidifier in accordance with an embodiment of the present invention; and
FIG. 5 is a schematic diagram of a device for processing the refrigerant content in a dehumidifier according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
In accordance with an embodiment of the present invention, there is provided an embodiment of a method for processing refrigerant content in a dehumidifier, where the steps illustrated in the flowchart of the drawings may be executed in a computer system, such as a set of computer executable instructions, and although a logical order is illustrated in the flowchart, in some cases, the steps illustrated or described may be executed in an order different from that shown.
Fig. 1 is a flowchart of a method for processing the refrigerant content in a dehumidifier according to an embodiment of the present invention, as shown in fig. 1, the method includes the following steps:
and step S102, obtaining the operation parameters of the dehumidifier.
Specifically, the above-mentioned operating parameters may include: the outdoor air inlet relative humidity, each dry bulb temperature, each wet bulb temperature, the exhaust temperature, the operation time, the current, the voltage, the power, the electric energy, and the like, but are not limited thereto, and the invention is not particularly limited thereto.
And step S104, analyzing the operation parameters by using a classification model, and determining the probability that the content of the refrigerant in the dehumidifier is various target contents.
Specifically, the refrigerant may be fluorine; the classification model may be a pre-trained neural network model, for example, a BP (Back Propagation) artificial neural network model, the target contents may be contents of refrigerants in a dehumidifier for training the classification model, the target contents may be 50%, 60%, 70%, and the like, and in the embodiment of the present invention, the details are described by taking the target contents as examples.
It should be noted that, in order to accurately determine the actual content of the refrigerant in the dehumidifier, the larger the number of the target content is, the better the target content is, but the larger the number of the target content is, the longer the operation time of the classification model is, and therefore, in the actual process, the number of the target content can be determined according to the accuracy and the time requirement.
And S106, determining the actual content of the refrigerant in the dehumidifier based on the probability of the various target contents and a preset loss value, wherein the preset loss value is used for representing the loss caused by judging the target content as the first refrigerant content by the classification model, and the first content is any one of the various target contents.
Specifically, the preset loss value may be a risk caused by prediction by the classification model, may be preset by expert experience, and may be a constant or a functional expression, for example, the loss value determined as 60% for 50% may be 0.1, and the loss value determined as 60% for 70% may be 0.6; the first content may be any one of a plurality of target contents, for example, three target contents are 50%, 60% and 70%, respectively, and when the first content is 50%, the preset loss value may include: judging 50% as a 50% loss value, 60% as a 50% loss value, and 70% as a 50% loss value; when the first content is 60%, the preset loss value may include: judging 50% as a 60% loss value, 60% as a 60% loss value, and 70% as a 60% loss value; when the first content is 70%, the preset loss value may include: the loss value was judged to be 70% for 50%, 70% for 60%, and 70% for 70%.
In an optional scheme, multiple target contents are taken as an example for explanation, when the fluorine content of the dehumidifier is predicted, the operation parameters of the dehumidifier can be input into a classification model, the classification model outputs the probability of each target content, in order to set out that the total error rate is rather increased and the total loss is reduced based on the loss consideration of different degrees caused by errors with different properties, the probability output by the classification model is input into a minimum risk decision function, and the minimum risk decision function calculates the content decision value of the refrigerant in the dehumidifier according to the probability output by the classification model and a preset loss value, that is, the actual content of the refrigerant in the dehumidifier is obtained.
In the embodiment of the invention, after the operation parameters of the dehumidifier are obtained, the operation parameters can be analyzed by using the classification model, the probability that the content of the refrigerant in the dehumidifier is various target contents is determined, and the actual content of the refrigerant in the dehumidifier is further determined based on the probability of the various target contents and the preset loss value, so that the prediction of the content of the refrigerant in the dehumidifier is realized. Because the minimum risk decision function is added after the probabilities of various target contents are output through the classification model, the actual content of the refrigerant in the dehumidifier is determined through the probabilities of the various target contents and the preset loss value, the loss caused by prediction errors of the classification algorithm is reduced, the processing accuracy and reliability of the refrigerant content in the dehumidifier are improved, and the technical problem of poor reliability of the processing method of the refrigerant content in the dehumidifier in the prior art is solved.
Optionally, determining an actual refrigerant content of the refrigerant in the dehumidifier based on the probability of the plurality of target contents and the preset loss value, including: obtaining a total loss value of the first content according to the probabilities of the various target contents and a preset loss value; obtaining a minimum total loss value of the total loss values of the plurality of first contents; and determining the first content corresponding to the minimum total loss value as the actual content.
In an optional scheme, multiple target contents are taken as an example for explanation, and due to the addition of the minimum risk decision function, a total loss value of the classification model for determining the content of the refrigerant in the dehumidifier as each target content can be obtained according to the probability of the multiple target contents output by the classification model and a preset loss value, and further, a final content decision value, namely, the actual content of the refrigerant in the dehumidifier is obtained by taking one item with the minimum total loss.
Optionally, obtaining a total loss value of the first content according to the probabilities of the various target contents and a preset loss value, including: obtaining the product of the probability of various target contents and a preset loss value to obtain a plurality of products; and obtaining the sum of the products to obtain the total loss value of the first content.
In an alternative scheme, taking multiple target contents as an example for description, for each target content in the multiple target contents, a product of the probability of each target content output by the classification model and a corresponding preset loss value may be calculated to obtain multiple products, and then the sum of all the products is calculated to obtain a total loss value of each target content. For example, there are three target contents, 50%, 60%, and 70%, respectively, and the classification model outputs a 50% probability, a 60% probability, and a 70% probability, and it may be determined that the preset loss value includes: the 50% loss value, the 60% loss value, and the 70% loss value are determined as 50% loss values, respectively, the product of the 50% probability and the 50% loss value, and the product of the 60% probability and the 60% loss value, which are determined as 50%, may be calculated, and then the three products may be added to obtain a total loss value of 50%.
It should be noted that the total loss value of each target content is calculated in the same manner, and the difference is only that the preset loss value is different.
Alternatively, the total loss value Y (α) of the first content is obtained by the following formulai|X)：
Wherein i, j is 1,2,3, …, n, n is the number of the target contents, alphaiDenotes a first content, p (ω)j| X) is the probability of the target content, ωjTo a target content, beta (. alpha.)iJ) is a predetermined loss value.
Specifically, the total loss value for each target content can be calculated by the above formula, where p (ω) isj| X) represents that the classification model judges the content X of the refrigerant in the dehumidifier as the target content omegajProbability of (a), beta (a)iJ) indicates that the content X is actually the target content omegajBut is judged as alpha by the classification modeliThe loss caused by the time.
It should be noted that the minimum risk decision function is Y (α)k|X)＝min(Y(αi| X)), wherein αkCan be used as the actual content of the refrigerant in the dehumidifier.
Optionally, the obtaining of the operating parameter of the dehumidifier includes: and after the dehumidifier normally operates for a first preset time, acquiring operation parameters every second preset time to obtain a plurality of operation parameters of the dehumidifier.
Specifically, the first preset time and the second preset time may be set as needed, for example, the first preset time may be n minutes, and the second preset time may be 3 minutes.
In an optional scheme, in the operation process of the dehumidifier, probability statistics decision logic can be added, namely, after the dehumidifier normally operates for n minutes, operation parameters are collected every 3 minutes, actual content is predicted until the dehumidifier is shut down, and probability statistics is carried out on multiple prediction results, so that final content output is obtained.
Optionally, determining the actual content of the refrigerant in the dehumidifier based on the probability of the various target contents and the preset loss value, including: determining a plurality of refrigerant contents of the refrigerant in the dehumidifier based on the probability of the plurality of target contents and a preset loss value; obtaining the proportion of various target contents according to the contents; obtaining the maximum proportion of the proportions of the various target contents; and determining the target content corresponding to the maximum proportion as the actual content.
In an optional scheme, through a probability statistics decision logic, after the operation parameters are collected once every 3 minutes, after the corresponding content is obtained through a classification model and a minimum risk decision function in a prediction mode, all prediction results can be counted, and the target content which occupies the largest proportion is used as the final content of the refrigerant in the dehumidifier to be output. For example, 10 predictions were obtained, 50%, 50%, 60%, 50%, 50%, 70%, respectively. As can be seen from the above, the ratio of 70% is 0.1, the ratio of 60% is 0.2, and the ratio of 50% is 0.7, and thus, the actual content of the refrigerant in the dehumidifier can be determined to be 50%.
Optionally, before analyzing the operation parameters by using the classification model and determining the probability that the content of the refrigerant in the dehumidifier is the multiple target contents, the method further includes: establishing an initial classification model; acquiring multiple groups of training sample data, wherein each group of training sample data comprises: the target content and the actual content of the refrigerant are the operation parameters of the dehumidifier with the target content under different working conditions; and training the initial classification model through multiple groups of training sample data to obtain a classification model.
In an optional scheme, under different working conditions, the dehumidifier with the actual content of the refrigerant being various target contents is operated, and the operation parameters of the dehumidifier are obtained through devices such as a sensor of the dehumidifier. And taking part of the acquired operation parameters as training samples, taking certain items of the acquired operation parameters as input of a classification algorithm, such as outdoor intake air relative humidity, dry-bulb temperature, exhaust temperature, operation time, current, voltage and the like, taking corresponding target content as a training label, and training an initial classification model through the training samples and the training label to obtain a trained classification model.
Optionally, after the initial classification model is trained through multiple sets of training sample data to obtain a classification model, the method further includes: acquiring multiple groups of test sample data, wherein each group of test sample data comprises: the target content and the actual content of the refrigerant are the operation parameters of the dehumidifier with the target content under different working conditions; testing the classification model through a plurality of groups of test sample data, and judging whether the output of the classification model meets a first preset condition; and if the output of the classification model does not meet the first preset condition, continuing training the classification model through multiple groups of training sample data until the output of the classification model meets the first preset condition.
Specifically, the first preset condition may be an expected requirement for determining whether the result predicted by the classification model is reasonable, for example, if the actual content of the refrigerant is 50%, and if the probability of 70% in the predicted result is the maximum, it is determined that the output of the classification model does not satisfy the first preset condition.
In an optional scheme, the remaining operating parameters in the acquired operating parameters may be used as test data to test the trained classification model, and whether the result output by the classification model meets an expected requirement is determined, where the expected requirement reflects that the accuracy of the output of the classification model meets a preset prediction target, and if not, the classification model continues to be trained until the expected requirement is met. For example, if the actual content of the refrigerant is 30% but the content output by the classification model is 50%, it is determined that the expected requirement is not met; and if the content of the output of the classification model is 40%, determining that the expected requirement is met.
Optionally, after determining that the output of the classification model satisfies the first preset condition, the method further includes: determining the actual content of the refrigerant in the dehumidifier based on the probability of various target contents output by the classification model and a preset loss value; judging whether the actual content meets a second preset condition or not; and if the actual content is determined not to meet the second preset condition, adjusting the preset loss value, and determining the actual content based on the probabilities of the various target contents and the adjusted preset loss value until the actual content meets the second preset condition.
Specifically, the second preset condition may be an expected requirement for determining whether the actual content predicted by the classification model and the minimum risk decision function is reasonable, for example, if the actual content of the refrigerant is less than 100% and greater than 30%, if the predicted actual content is less than 30%, it is determined that the actual content does not satisfy the second preset condition.
In an optional scheme, if the actual content obtained through the classification model and the minimum risk decision function is unreasonable, that is, does not satisfy the second preset condition, it indicates that the preset loss value needs to be adjusted, and the actual content is predicted again according to the adjusted preset loss value until the actual content satisfies the second preset condition.
Optionally, before determining that the actual content satisfies the second preset condition, the method further includes: and storing the classification model and the preset loss value to a server and a controller of the dehumidifier.
Specifically, the server may be a cloud server, and the controller may be a main control chip of the dehumidifier.
In an optional scheme, the classification algorithm, the minimum risk decision function and the probability statistics decision logic can be written into a control chip and a cloud server of the dehumidifier, and when the dehumidifier with wireless communication runs, the running parameters of the dehumidifier are uploaded to the chip or the cloud server for processing. If the fluorine lack condition of the dehumidifier occurs, the fluorine lack condition can be reported to the dehumidifier in time.
Optionally, before obtaining the operating parameters of the dehumidifier, the method further includes: judging whether the dehumidifier has a networking function or not; if the dehumidifier has the networking function, the operation parameters are sent to the slave server, the actual content returned by the server is received, or the operation parameters are processed through a controller of the dehumidifier to obtain the actual content; and if the dehumidifier does not have the networking function, processing the operation parameters through a controller of the dehumidifier to obtain the actual content.
In an optional scheme, after the operation parameters of the dehumidifier are obtained, whether the dehumidifier has a networking function can be judged, if yes, the operation parameters can be sent to a cloud server through a network for processing, so that the actual content of the dehumidifier is obtained, or the operation parameters can be processed through a controller, so that the actual content of the dehumidifier is obtained; if not, the operation parameters can be directly processed by the controller to obtain the actual content of the dehumidifier.
Optionally, after determining the actual content of the refrigerant in the dehumidifier based on the probabilities of the various target contents and the preset loss value, the method further includes: judging whether the actual content is lower than a preset value; and if the actual content is lower than the preset value, generating alarm prompt information or controlling the dehumidifier to execute preset action, wherein the alarm prompt information is used for prompting that the actual content is lower than the preset value.
Specifically, the preset value may be the minimum content for judging that the dehumidifier lacks fluorine, and may be set according to the actual use requirement, for example, may be set to 60%. The preset action may be turning off or adjusting the operating parameters of the dehumidifier to reduce the fluorine usage, which is not specifically limited in the present invention.
In an optional scheme, after the actual content of the refrigerant in the dehumidifier is obtained through prediction, whether the dehumidifier lacks the refrigerant or not can be judged by comparing the actual content with a preset value, namely, whether the dehumidifier lacks fluorine or not is judged, if the actual content is less than or equal to the preset value, the dehumidifier lacks fluorine is determined, alarm prompt information needs to be generated and displayed by a display screen of the dehumidifier, and a user is prompted that the dehumidifier lacks fluorine; or the dehumidifier can be directly controlled to be turned off.
Fig. 2 is a schematic structural diagram of a method for processing the refrigerant content in an optional dehumidifier according to an embodiment of the present invention, fig. 3 is a flowchart of optional classification model training and preset loss value adjustment according to an embodiment of the present invention, fig. 4 is a flowchart of a method for processing the refrigerant content in an optional dehumidifier according to an embodiment of the present invention, and a preferred embodiment of the present invention will be described in detail with reference to fig. 2 to fig. 4 by taking fluorine content prediction in the dehumidifier as an example.
As shown in fig. 2, the collected operation parameters may be input into an input layer of the classification model, features are extracted through a hidden layer, and probabilities of various fluorine contents are output through an output layer, the probabilities output by the classification model are input into a minimum risk decision function, and finally, a final fluorine content type is output, and when the dehumidifier operates, a statistical decision is performed on the fluorine content type output by the minimum risk decision function through a probability statistical decision logic, so that the fluorine content of the dehumidifier is obtained.
As shown in fig. 3, under different working conditions, operating the dehumidifier with different n types of refrigerant contents, obtaining operating parameters of the dehumidifier through a sensor of the dehumidifier and other devices, selecting a part of data as a training sample, using a part of the obtained operating parameters as input of an artificial neural network algorithm, using the corresponding refrigerant content as a training label, and modeling and training through the artificial neural network algorithm to obtain a trained artificial neural network model; and the other part of data is used as a test sample, a trained artificial neural network algorithm is tested, whether the output meets the expected requirement is judged, and if not, the training is continued until the output meets the expected requirement. The loss beta (alpha) misjudged as the nth class is preset by expert experienceiAnd j (i, j is 1,2,3, …, n), adding a minimum risk decision function after an artificial neural network algorithm to predict the refrigerant proportion of the training sample, counting the prediction result, and if so, adjusting the loss until the prediction result is reasonable. Setting probability statistic decision logic, and writing an algorithm (including an artificial neural network algorithm and a minimum risk decision function) and the probability statistic decision logic into a chip or a cloud.
Assuming a total of 4 tags, and predicting fluorine deficiency starting from a value smaller than tag 2, the misjudgment loss between tags can be as shown in table 1 below:
TABLE 1
| Loss value | Is judged to be 1 | Is judged to be 2 | Is judged to be 3 | Judged as 4 |
| Label 1 | 0 | 4-7 | 37-40 | 38-42 |
| Label 2 | 4-7 | 0 | 14-18 | 28-32 |
| Label 3 | 17-21 | 15-20 | 0 | 1-4 |
| Label 4 | 38-42 | 28-32 | 1-5 | 0 |
Taking 5 misjudgment samples in the primary test result, adopting minimum risk decision, wherein the difference is shown in the following two tables 2 and 3, wherein the table 2 shows the result obtained by directly carrying out decision according to the probability obtained by the artificial neural network algorithm before the minimum risk decision is adopted, and the table 3 shows the result obtained after the minimum risk decision is adopted. As shown in table 2, the other 4 samples except for samples 2 and 4 in 5 samples all have error reporting, and as shown in table 3, after the minimum risk function is adopted, the prediction result is more reasonable, and 5 samples do not have error reporting.
TABLE 2
TABLE 3
As shown in fig. 4, when the dehumidifier runs, an operation parameter is generated, whether the dehumidifier has a networking function is judged, if yes, the operation parameter is input into a chip or a cloud as an algorithm input, and the fluorine content is predicted through the algorithm and probability statistics decision logic; if not, the operating parameters are directly used as the input of the algorithm on the chip, and the fluorine content is predicted through the algorithm and probability statistics decision logic. After the refrigerant quantity of the dehumidifier is predicted, the result is returned to the dehumidifier controller for judgment, and whether fluorine is lacking is judged; if the fluorine is not available, the display screen of the dehumidifier is forecast or is forecast in other forms, and some control is performed, such as display screen forecast and shutdown (not limited), if the fluorine is not available, the process is executed again when the dehumidifier is started next time.
According to the scheme, the decision scheme for predicting the fluorine content of the dehumidifier based on the classification algorithm is provided, the minimum risk decision function is added after the classification algorithm, and the probability statistics decision logic is added when the dehumidifier runs, so that the risk, namely loss, caused by prediction errors of the classification algorithm is reduced. The fluorine content of the dehumidifier is more accurate and reasonable than the logic judgment given by a control technician, and the dehumidifier has strong generalization capability.
According to an embodiment of the present invention, an embodiment of a device for processing refrigerant content in a dehumidifier is provided.
Fig. 5 is a schematic diagram of a device for processing the refrigerant content in a dehumidifier according to an embodiment of the present invention, as shown in fig. 5, the device includes:
and the obtaining module 52 is used for obtaining the operating parameters of the dehumidifier.
Specifically, the above-mentioned operating parameters may include: the outdoor air inlet relative humidity, each dry bulb temperature, each wet bulb temperature, the exhaust temperature, the operation time, the current, the voltage, the power, the electric energy, and the like, but are not limited thereto, and the invention is not particularly limited thereto.
The first determining module 54 is configured to analyze the operation parameters by using the classification model, and determine the probability that the content of the refrigerant in the dehumidifier is the multiple target contents.
Specifically, the refrigerant may be fluorine; the classification model may be a pre-trained neural network model, for example, a BP artificial neural network model, the target contents may be contents of refrigerants in a dehumidifier for training the classification model, and the target contents may be 50%, 60%, 70%, and the like.
It should be noted that, in order to accurately determine the actual content of the refrigerant in the dehumidifier, the larger the number of the target content is, the better the target content is, but the larger the number of the target content is, the longer the operation time of the classification model is, and therefore, in the actual process, the number of the target content can be determined according to the accuracy and the time requirement.
The second determining module 56 is configured to determine an actual content of the refrigerant in the dehumidifier based on the probabilities of the multiple target contents and a preset loss value, where the preset loss value is used to represent a loss caused by the classification model determining the target content as the first refrigerant content, and the first content is any one of the multiple target contents.
Specifically, the preset loss value may be a risk caused by prediction by the classification model, may be preset by expert experience, and may be a constant or a functional expression, for example, the loss value determined as 60% for 50% may be 0.1, and the loss value determined as 60% for 70% may be 0.6; the first content may be any one of a plurality of target contents, for example, three target contents are 50%, 60% and 70%, respectively, and when the first content is 50%, the preset loss value may include: judging 50% as a 50% loss value, 60% as a 50% loss value, and 70% as a 50% loss value; when the first content is 60%, the preset loss value may include: judging 50% as a 60% loss value, 60% as a 60% loss value, and 70% as a 60% loss value; when the first content is 70%, the preset loss value may include: the loss value was judged to be 70% for 50%, 70% for 60%, and 70% for 70%.
In an optional scheme, multiple target contents are taken as an example for explanation, when the fluorine content of the dehumidifier is predicted, the operation parameters of the dehumidifier can be input into a classification model, the classification model outputs the probability of each target content, in order to set out that the total error rate is rather increased and the total loss is reduced based on the loss consideration of different degrees caused by errors with different properties, the probability output by the classification model is input into a minimum risk decision function, and the minimum risk decision function calculates the content decision value of the refrigerant in the dehumidifier according to the probability output by the classification model and a preset loss value, that is, the actual content of the refrigerant in the dehumidifier is obtained.
In the embodiment of the invention, after the operation parameters of the dehumidifier are obtained, the operation parameters can be analyzed by using the classification model, the probability that the content of the refrigerant in the dehumidifier is various target contents is determined, and the actual content of the refrigerant in the dehumidifier is further determined based on the probability of the various target contents and the preset loss value, so that the prediction of the content of the refrigerant in the dehumidifier is realized. Because the minimum risk decision function is added after the probabilities of various target contents are output through the classification model, the actual content of the refrigerant in the dehumidifier is determined through the probabilities of the various target contents and the preset loss value, the loss caused by prediction errors of the classification algorithm is reduced, the processing accuracy and reliability of the refrigerant content in the dehumidifier are improved, and the technical problem of poor reliability of the processing method of the refrigerant content in the dehumidifier in the prior art is solved.
According to an embodiment of the present invention, an embodiment of a storage medium is provided, where the storage medium includes a stored program, and when the program runs, a device in which the storage medium is located is controlled to execute the above method for processing the content of the media in the dehumidifier.
According to an embodiment of the present invention, an embodiment of a processor is provided, where the processor is configured to run a program, and the program executes the method for processing the refrigerant content in the dehumidifier when running.
In the above embodiments of the present invention, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
In the embodiments provided in the present application, it should be understood that the disclosed technology can be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units may be a logical division, and in actual implementation, there may be another division, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, units or modules, and may be in an electrical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (15)
1. A method for processing the content of refrigerant in a dehumidifier is characterized by comprising the following steps:
acquiring the operating parameters of the dehumidifier;
analyzing the operation parameters by using a classification model to determine the probability that the content of the refrigerant in the dehumidifier is various target contents;
and determining the actual content of the refrigerant in the dehumidifier through a minimum risk decision function based on the probability of the various target contents and a preset loss value, wherein the preset loss value is used for representing the loss caused by judging the target contents as first contents by the classification model, and the first contents are any one of the various target contents.
2. The method as claimed in claim 1, wherein determining the actual content of the refrigerant in the dehumidifier through a minimum risk decision function based on the probability of the plurality of target contents and a preset loss value comprises:
obtaining a total loss value of the first content according to the probabilities of the various target contents and the preset loss value;
obtaining a minimum total loss value of the total loss values of the plurality of first contents;
and determining the first content corresponding to the minimum total loss value as the actual content.
3. The method of claim 2, wherein obtaining the total loss value of the first content according to the probabilities of the plurality of target contents and the preset loss value comprises:
obtaining the product of the probability of the various target contents and the preset loss value to obtain a plurality of products;
and obtaining the sum of the products to obtain the total loss value of the first content.
4. The method according to claim 3, wherein the total loss value Y (α) of the first content is obtained by the following formulai|X)：
Wherein i, j is 1,2,3iRepresents said first content, said p (ω)j| X) is to determine the content X of the refrigerant in the dehumidifierIs determined as the target content omegajProbability of, said ωjTo said target content, beta (. alpha.)iJ) is the preset loss value, and X is the content of the refrigerant in the dehumidifier.
5. The method of claim 1, wherein obtaining operating parameters of a dehumidifier comprises:
and after the dehumidifier normally operates for a first preset time, acquiring the operating parameters every second preset time to obtain a plurality of operating parameters of the dehumidifier.
6. The method as claimed in claim 5, wherein determining the actual content of the refrigerant in the dehumidifier through a minimum risk decision function based on the probability of the plurality of target contents and a preset loss value comprises:
determining a plurality of contents of the refrigerant in the dehumidifier through a minimum risk decision function based on the probabilities of the plurality of target contents and a preset loss value;
obtaining the proportion of the plurality of target contents according to the plurality of contents;
obtaining the maximum proportion in the proportions of the various target contents;
and determining the target content corresponding to the maximum proportion as the actual content.
7. The method of claim 1, wherein before analyzing the operating parameters using a classification model to determine a probability that the refrigerant content of the dehumidifier is a plurality of target contents, the method further comprises:
establishing an initial classification model;
acquiring multiple groups of training sample data, wherein each group of training sample data comprises: the target content and the actual content of the refrigerant are the operating parameters of the dehumidifier with the target content under different working conditions;
and training the initial classification model through the multiple groups of training sample data to obtain the classification model.
8. The method of claim 7, wherein after training the initial classification model through the plurality of sets of training sample data to obtain the classification model, the method further comprises:
acquiring multiple groups of test sample data, wherein each group of test sample data comprises: the target content and the actual content of the refrigerant are the operating parameters of the dehumidifier with the target content under different working conditions;
testing the classification model through the multiple groups of test sample data, and judging whether the output of the classification model meets a first preset condition, wherein the first preset condition is an expected requirement for judging whether a result obtained by prediction of the classification model is reasonable;
if the output of the classification model does not meet the first preset condition, continuing to train the classification model through the multiple groups of training sample data until the output of the classification model meets the first preset condition.
9. The method of claim 8, wherein after determining that the output of the classification model satisfies the first preset condition, the method further comprises:
determining the actual content of the refrigerant in the dehumidifier through the minimum risk decision function based on the probability of various target contents output by the classification model and a preset loss value;
judging whether the actual content meets a second preset condition, wherein the second preset condition is an expected requirement for judging whether the actual content predicted by the classification model and the minimum risk decision function is reasonable;
and if the actual content is determined not to meet the second preset condition, adjusting the preset loss value, and determining the actual content based on the probabilities of the various target contents and the adjusted preset loss value until the actual content meets the second preset condition.
10. The method according to claim 9, wherein before determining that the actual content satisfies the second preset condition, the method further comprises:
and storing the classification model and the preset loss value to a server and a controller of the dehumidifier.
11. The method of claim 1, wherein after obtaining the operating parameters of the dehumidifier, the method further comprises:
judging whether the dehumidifier has a networking function or not;
if the dehumidifier has the networking function, the operation parameters are sent to a slave server, the actual content returned by the server is received, or the operation parameters are processed through a controller of the dehumidifier to obtain the actual content;
and if the dehumidifier does not have the networking function, processing the operation parameters through a controller of the dehumidifier to obtain the actual content.
12. The method as claimed in claim 1, wherein after determining the actual content of the refrigerant in the dehumidifier through a minimum risk decision function based on the probabilities of the plurality of target contents and a preset loss value, the method further comprises:
judging whether the actual content is lower than a preset value or not;
and if the actual content is lower than the preset value, generating alarm prompt information or controlling the dehumidifier to execute preset action, wherein the alarm prompt information is used for prompting that the actual content is lower than the preset value.
13. A device for processing the content of refrigerant in a dehumidifier is characterized by comprising:
the acquisition module is used for acquiring the operating parameters of the dehumidifier;
the first determination module is used for analyzing the operation parameters by using a classification model and determining the probability that the content of the refrigerant in the dehumidifier is various target contents;
and the second determining module is used for determining the actual content of the refrigerant in the dehumidifier through a minimum risk decision function based on the probabilities of the various target contents and a preset loss value, wherein the preset loss value is used for representing the loss caused by judging the target content as the first content by the classification model, and the first content is any one of the various target contents.
14. A storage medium, characterized in that the storage medium comprises a stored program, wherein when the program runs, the device where the storage medium is located is controlled to execute the method for processing the content of the refrigerant in the dehumidifier according to any one of claims 1 to 12.
15. A processor, wherein the processor is configured to run a program, and the program is configured to execute the method for processing the refrigerant content in the dehumidifier according to any one of claims 1 to 12 when running.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810427239.5A CN110454904B (en) | 2018-05-07 | 2018-05-07 | Method and device for processing refrigerant content in dehumidifier |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810427239.5A CN110454904B (en) | 2018-05-07 | 2018-05-07 | Method and device for processing refrigerant content in dehumidifier |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110454904A CN110454904A (en) | 2019-11-15 |
| CN110454904B true CN110454904B (en) | 2020-12-11 |
Family
ID=68471954
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810427239.5A Active CN110454904B (en) | 2018-05-07 | 2018-05-07 | Method and device for processing refrigerant content in dehumidifier |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110454904B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPWO2015004747A1 (en) * | 2013-07-10 | 2017-02-23 | 三菱電機株式会社 | Refrigeration cycle equipment |
| CN107655162A (en) * | 2017-08-21 | 2018-02-02 | 珠海格力电器股份有限公司 | Refrigerant detection method, device, dehumidifier and equipment |
| CN107730000A (en) * | 2017-09-15 | 2018-02-23 | 珠海格力电器股份有限公司 | Coolant leakage detection method and device for air-conditioning |
-
2018
- 2018-05-07 CN CN201810427239.5A patent/CN110454904B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPWO2015004747A1 (en) * | 2013-07-10 | 2017-02-23 | 三菱電機株式会社 | Refrigeration cycle equipment |
| CN107655162A (en) * | 2017-08-21 | 2018-02-02 | 珠海格力电器股份有限公司 | Refrigerant detection method, device, dehumidifier and equipment |
| CN107730000A (en) * | 2017-09-15 | 2018-02-23 | 珠海格力电器股份有限公司 | Coolant leakage detection method and device for air-conditioning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110454904A (en) | 2019-11-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7188255B2 (en) | LEARNING METHOD, LEARNING PROGRAM AND LEARNING DEVICE | |
| CN111177714B (en) | Abnormal behavior detection method and device, computer equipment and storage medium | |
| AU2016286280A1 (en) | Combined method for detecting anomalies in a water distribution system | |
| CN106845526B (en) | A kind of relevant parameter Fault Classification based on the analysis of big data Fusion of Clustering | |
| CN105488539B (en) | The predictor method and device of the generation method and device of disaggregated model, power system capacity | |
| CN106446317A (en) | Mathematic model-based sealed relay storage life prediction method | |
| Sun et al. | Optimization of support vector regression model based on outlier detection methods for predicting electricity consumption of a public building WSHP system | |
| CN111174370A (en) | Fault detection method and device, storage medium and electronic device | |
| CN109726767A (en) | A kind of perceptron network data classification method based on AdaBoost algorithm | |
| CN113313280B (en) | Cloud platform inspection method, electronic equipment and nonvolatile storage medium | |
| JP2019121162A (en) | Monitoring device, monitoring method, and monitoring program | |
| Yan et al. | A Decentralized Boltzmann-machine-based fault diagnosis method for sensors of Air Handling Units in HVACs | |
| CN112463773A (en) | Data quality determination method and device | |
| WO2019214230A1 (en) | Air conditioner defrosting method and device | |
| CN111984442A (en) | Method and device for detecting abnormality of computer cluster system, and storage medium | |
| CN113705688A (en) | Method and system for detecting abnormal electricity utilization behavior of power consumer | |
| CN111444075A (en) | Method for automatically discovering key influence indexes | |
| CN113098912B (en) | User account abnormity identification method and device, electronic equipment and storage medium | |
| CN110361982A (en) | A kind of home equipment control method, storage medium and smart home device | |
| CN111738348B (en) | Power data anomaly detection method and device | |
| CN110454904B (en) | Method and device for processing refrigerant content in dehumidifier | |
| WO2021079485A1 (en) | Generation method, generation program, and information processing device | |
| US20220222580A1 (en) | Deterioration detection method, non-transitory computer-readable storage medium, and information processing device | |
| CN111504673A (en) | Fault diagnosis method and system for water chilling unit and air conditioner | |
| CN111159251A (en) | Method and device for determining abnormal data |
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 |