CN113587362A - Abnormity detection method and device and air conditioning system - Google Patents

Abnormity detection method and device and air conditioning system Download PDF

Info

Publication number
CN113587362A
CN113587362A CN202110875546.1A CN202110875546A CN113587362A CN 113587362 A CN113587362 A CN 113587362A CN 202110875546 A CN202110875546 A CN 202110875546A CN 113587362 A CN113587362 A CN 113587362A
Authority
CN
China
Prior art keywords
air conditioning
conditioning system
encoder
parameters
target
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110875546.1A
Other languages
Chinese (zh)
Inventor
吴斌
范波
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Midea Group Co Ltd
GD Midea Heating and Ventilating Equipment Co Ltd
Guangdong Midea HVAC Equipment Co Ltd
Original Assignee
Midea Group Co Ltd
GD Midea Heating and Ventilating Equipment Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Midea Group Co Ltd, GD Midea Heating and Ventilating Equipment Co Ltd filed Critical Midea Group Co Ltd
Priority to CN202110875546.1A priority Critical patent/CN113587362A/en
Publication of CN113587362A publication Critical patent/CN113587362A/en
Pending legal-status Critical Current

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
    • F24F11/32—Responding to malfunctions or emergencies
    • F24F11/38—Failure diagnosis
    • 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
    • F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
    • F24F11/63—Electronic processing
    • F24F11/64—Electronic processing using pre-stored data
    • 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
    • F24F11/88—Electrical aspects, e.g. circuits

Abstract

The application is suitable for the technical field of air conditioners, and provides an abnormality detection method, an abnormality detection device and an air conditioning system, wherein the abnormality detection method comprises the following steps: acquiring target parameters, wherein the target parameters comprise operation parameters of the air conditioning system; inputting the target parameter into a pre-constructed self-encoder to obtain an output value of the pre-constructed self-encoder; determining whether the air conditioning system is abnormal or not according to the target parameter, the output value and a preset alarm threshold range; and if the updating condition is met, updating the pre-constructed self-encoder and the alarm threshold according to the acquired target parameter. By the method, a more accurate abnormity judgment result can be obtained.

Description

Abnormity detection method and device and air conditioning system
Technical Field
The application belongs to the technical field of air conditioners, and particularly relates to an abnormality detection method and device, an air conditioning system and a computer readable storage medium.
Background
In the operation process of the air conditioner, various faults existing in the air conditioning system generally need to be detected so as to process the faults in time, and the energy consumption of the unit can be effectively reduced and the service life of the unit can be prolonged by processing the faults in time.
When the air conditioning system is detected to be abnormal, an alarm is generated. At present, after obtaining detection data of an air conditioning system, the detection data is input into a preset fault detection model based on Principal Component Analysis (PCA), and a detection result output by the detection model is obtained. However, it is still difficult to obtain accurate detection results by using this method.
Therefore, it is necessary to provide a new method to solve the above technical problems.
Disclosure of Invention
The embodiment of the application provides an abnormality detection method and device and an air conditioning system, and can solve the problem that whether the air conditioning system is abnormal or not is difficult to accurately judge by the existing method.
In a first aspect, an embodiment of the present application provides an anomaly detection method, applied to an air conditioning system, including:
acquiring target parameters, wherein the target parameters comprise operation parameters of the air conditioning system;
inputting the target parameter into a pre-constructed self-encoder to obtain an output value of the pre-constructed self-encoder;
determining whether the air conditioning system is abnormal or not according to the target parameter, the output value and a preset alarm threshold range;
and if the updating condition is met, updating the pre-constructed self-encoder according to the acquired target parameter.
In a second aspect, an embodiment of the present application provides an abnormality detection apparatus, which is applied to an air conditioning system, and includes:
the target parameter acquisition module is used for acquiring target parameters, and the target parameters comprise operation parameters of the air conditioning system;
the output value determining module is used for inputting the target parameter into a pre-constructed self-encoder to obtain the output value of the pre-constructed self-encoder;
the abnormity judging module is used for determining whether the air conditioning system is abnormal or not according to the target parameter, the output value and a preset alarm threshold range;
and the self-encoder updating module is used for updating the pre-constructed self-encoder according to the acquired target parameters if the updating conditions are met.
In a third aspect, an embodiment of the present application provides an air conditioning system, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the method according to the first aspect when executing the computer program.
In a fourth aspect, the present application provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the computer program implements the method according to the first aspect.
In a fifth aspect, embodiments of the present application provide a computer program product, which, when run on an air conditioning system, causes the air conditioning system to perform the method described in the first aspect.
Compared with the prior art, the embodiment of the application has the advantages that:
in the embodiment of the application, since the target parameter comprises the operation parameter of the air conditioning system, after the target parameter is input into the self-encoder, the output value of the self-encoder is also related to the operation parameter, and therefore, whether the air conditioning system is abnormal or not can be judged according to the output value, the preset alarm threshold range and the target parameter. Meanwhile, compared with a fault detection model based on PCA, the self-encoder can adapt to an air conditioning system with higher complexity, so that after the characteristics of the target parameters are learned through the self-encoder of the embodiment of the application, more accurate output values can be obtained, and further, when abnormality judgment is carried out according to the more accurate output values, more accurate abnormality judgment results can be obtained. In addition, the self-encoder of the embodiment of the application can also be used for updating on line by combining the acquired target parameters, so that the anomaly detection method provided by the embodiment of the application can be suitable for air conditioning systems under different working conditions and/or different units, namely, even if the anomaly detection is carried out on the air conditioning systems under different working conditions and/or different units, an accurate anomaly judgment result can be obtained.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings used in the embodiments or the description of the prior art will be briefly described below.
Fig. 1 is a flowchart of a first anomaly detection method according to an embodiment of the present application;
fig. 2 is a schematic structural diagram of a self-encoder according to an embodiment of the present application;
FIG. 3 is a flow chart of a second anomaly detection method provided by an embodiment of the present application;
FIG. 4 is a graphical illustration of an alarm threshold range provided by an embodiment of the present application;
FIG. 5 is a flow chart of a third method for anomaly detection provided by an embodiment of the present application;
fig. 6 is a schematic structural diagram of an abnormality detection apparatus according to an embodiment of the present application;
fig. 7 is a schematic structural diagram of an air conditioning system according to an embodiment of the present application.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular system structures, techniques, etc. in order to provide a thorough understanding of the embodiments of the present application. It will be apparent, however, to one skilled in the art that the present application may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail.
It will be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It should also be understood that the term "and/or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.
Reference throughout this specification to "one embodiment" or "some embodiments," or the like, means that a particular feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of the present application. That is, the appearances of the phrases "in one embodiment," "in some embodiments," "in other embodiments," or the like, appearing in various places throughout the specification are not necessarily all referring to the same embodiment, but rather "one or more but not all embodiments" unless specifically stated otherwise.
The first embodiment is as follows:
in the existing detection method, detection data of the air conditioning system is input into a preset fault detection model based on PCA, and whether the air conditioning system is abnormal or not is determined according to a detection result output by the detection model. The fault detection model based on the PCA is obtained by training aiming at a fixed working condition and/or a fixed unit, so the fault detection model based on the PCA is a static detection model, namely the fault detection model based on the PCA is difficult to adapt to the abnormal detection of the air conditioning system with the changed working condition and the abnormal detection of the air conditioning system with the changed unit, namely the fault detection model based on the PCA is difficult to obtain an accurate detection result. In addition, the inventor of the present application can know through analysis that, because the complexity of the existing air conditioning system is getting bigger and bigger, even if the working condition or the unit of the air conditioning system is not changed, when the fault detection model based on the PCA is adopted to detect the abnormality of the air conditioning system, it is difficult to obtain an accurate detection result.
In order to solve the above technical problem, an embodiment of the present application provides an anomaly detection method, in which a pre-constructed self-encoder is used to process a target parameter including an operation parameter of an air conditioning system to obtain a corresponding output value, and finally, whether the air conditioning system is anomalous or not is determined according to the output value, a preset alarm threshold range and the target parameter. Because the self-encoder can adapt to an air conditioning system with higher complexity, the self-encoder based on the pre-construction can obtain more accurate output values, and further more accurate abnormal judgment results are obtained. In addition, the pre-constructed self-encoder of the embodiment of the application can also be updated on line by combining the acquired target parameters, so that the anomaly detection method provided by the embodiment of the application can adapt to air conditioning systems under different working conditions and/or different units, namely, even if the anomaly detection is carried out on the air conditioning systems under different working conditions and/or different units, an accurate detection result can be obtained.
The abnormality detection method provided by the embodiment of the present application is described below with reference to the drawings.
Fig. 1 is a flowchart illustrating a first anomaly detection method provided in an embodiment of the present application, which is applied to an air conditioning system, where the air conditioning system may be a system adopted by an air conditioner with single air conditioner (i.e., one air conditioner outdoor host only carries one air conditioner indoor end) or a system adopted by an air conditioner with multiple air conditioners (i.e., one air conditioner outdoor host plus two or more air conditioner indoor ends), and details are as follows:
and step S11, acquiring target parameters, wherein the target parameters comprise the operation parameters of the air conditioning system.
In this embodiment, when the air conditioning system is in operation, the operation parameters generated during the operation of the air conditioning system, such as the compressor frequency, the condensation temperature, the evaporation temperature, the exhaust pressure, the suction pressure, and the like, are obtained.
In some embodiments, the target parameter further includes an environmental parameter, such as, for example, including outdoor temperature.
In some embodiments, since the obtained target parameter is to be used for detecting whether an abnormality occurs in the air conditioning system, in order to reduce the amount of data to be processed subsequently, the target parameter may be obtained after an interval duration arrives, where the interval duration is in units of seconds. For example, assuming that the duration of the interval is 5 seconds, the corresponding target parameter is acquired every 5 seconds. The target parameters corresponding to all the moments are not acquired, so that the number of the acquired target parameters can be effectively reduced, and the resources of the air conditioning system can be effectively saved. In addition, because the unit of the interval duration is 'second', namely the interval duration is short, even if the target parameters at each moment are not acquired, whether the air conditioning system is abnormal or not can be timely judged according to the subsequently acquired target parameters.
And step S12, inputting the target parameters into the pre-constructed self-encoder to obtain the output value of the pre-constructed self-encoder.
The pre-constructed self-encoder (hereinafter referred to as self-encoder) is a kind of artificial neural network used in semi-supervised learning and unsupervised learning.
In this embodiment, the structure of the self-encoder is a symmetric structure, such as the encoder in the self-encoder and the decoder in the self-encoder are symmetric. The number of hidden layers corresponding to the encoder and the decoder can be determined according to actual conditions. In some scenarios, assuming that the number of hidden layers corresponding to the encoder and the decoder are both 1, the structure of the self-encoder of this embodiment may be as shown in fig. 2. In fig. 2, the encoder includes an input layer and a first hidden layer; and the decoder comprises a third hidden layer and an output layer. As can be seen in fig. 2, the encoder and decoder of the self-encoder are symmetric.
In the present embodiment, the acquired target parameter is used as an input of the self-encoder, and after the characteristics of the input target parameter are learned by the self-encoder, a corresponding numerical value (i.e., the output value) is output.
And step S13, determining whether the air conditioning system is abnormal or not according to the target parameter, the output value and the preset alarm threshold range.
In this embodiment, a certain operation is performed on the target parameter and the output value, for example, a subtraction operation is performed on the target parameter and the output value, or a division operation is performed on the target parameter and the output value, so as to obtain a corresponding calculation result. And comparing the calculation result with a preset alarm threshold range, if the calculation result is within the preset alarm threshold range, judging that the air conditioning system is abnormal, otherwise, if the calculation result is not within the preset alarm threshold range, judging that the air conditioning system is abnormal.
And step S14, if the updating condition is met, updating the pre-constructed self-encoder according to the acquired target parameters.
In this embodiment, the update condition may be set according to an actual situation, for example, the update condition may be any of the following: the set update time is up, any target parameter (equivalent to real-time update) is acquired, and an update instruction sent by a user is received. For example, if the update condition is that the set update time is reached, the pre-constructed self-encoder will be automatically updated according to the acquired target parameters when the set update time is reached.
In some embodiments, if the pre-constructed self-encoder starts to be updated without acquiring any target parameter (i.e. the pre-constructed self-encoder is not updated in real time), the target parameter for updating the pre-constructed self-encoder may be: target parameters obtained during a time period after the last update action was performed and before the current update action was performed. For example, assume that the point in time when the last update action was performed is 12 on 6 months, 4 days: 00, the time point for executing the current update action is 6 months, 10 days, 12: 00, 6 months, 4 days 12: month 01 to 6, day 10 11: the target parameters acquired during this time period 59 are used to update the pre-constructed self-encoder. All the obtained target parameters are used for updating the pre-constructed self-encoder, namely, samples are added, so that the method is beneficial to obtaining more accurate updated self-encoder, and further more accurate alarm threshold value is obtained.
In the embodiment of the application, since the target parameter comprises the operation parameter of the air conditioning system, after the target parameter is input into the self-encoder, the output value of the self-encoder is also related to the operation parameter, and therefore, whether the air conditioning system is abnormal or not can be judged according to the output value, the preset alarm threshold range and the target parameter. Meanwhile, compared with a fault detection model based on PCA, the self-encoder can adapt to an air conditioning system with higher complexity, so that after the characteristics of the target parameters are learned through the self-encoder of the embodiment of the application, more accurate output values can be obtained, and further, when abnormality judgment is carried out according to the more accurate output values, more accurate abnormality judgment results can be obtained. In addition, the self-encoder of the embodiment of the application can also be used for updating on line by combining the acquired target parameters, so that the anomaly detection method provided by the embodiment of the application can be suitable for air conditioning systems under different working conditions and/or different units, namely, even if the anomaly detection is carried out on the air conditioning systems under different working conditions and/or different units, an accurate anomaly judgment result can be obtained.
Example two:
fig. 3 shows a flowchart of a second anomaly detection method provided in an embodiment of the present application. In this embodiment, steps S33 to S36 are the same as steps S11 to S14 of the first embodiment, and are not repeated here. In addition, the present embodiment adds a step of how to generate a pre-constructed self-encoder on the basis of the first embodiment, which is detailed as follows:
and step S31, determining structural parameters according to the complexity of the air conditioning system.
In this embodiment, the complexity of the air conditioning system is related to the structure of the air conditioner itself and/or the function of the air conditioner, for example, the complexity of the air conditioning system corresponding to the multi-split air conditioner is higher than that of the single-split air conditioner. For example, air conditioning systems for air conditioners with multiple functions have a higher complexity than air conditioners with a single function.
In this embodiment, the structural parameters are structural parameters of an auto-encoder. Since the more complex the air conditioning system, the more variables are input, the more the number of neuron nodes per layer is caused. That is, when the air conditioning system is more complex, the corresponding structural parameters are more, and conversely, the corresponding structural parameters are less. Taking an air conditioning system corresponding to a multi-split air conditioner as an example, assuming that there are 50 input variables (or characteristics), the number of nodes in the first layer may be 50, and the subsequently obtained network structure corresponding to the pre-constructed self-encoder may be 50-24-12-24-50, where one number represents one layer of network, that is, in the above-listed network structure, the number of network layers is 5, and the size of the number in the above-listed network structure represents the number of neuron nodes in the current network layer.
And step S32, acquiring historical target parameters, and training the self-encoder constructed based on the structural parameters according to the historical target parameters to obtain a pre-constructed self-encoder.
The above-mentioned history target parameter refers to the target parameter acquired before step S33 is executed.
In some embodiments, the acquired historical target parameters are target parameters corresponding to air conditioning systems similar to or identical to the subsequent air conditioning system that needs to be determined whether an abnormality exists. Since the air conditioning system where the historical target parameters are located is the same as or similar to the air conditioning system where the target parameters obtained in the subsequent step S33 are located, the historical target parameters obtained in the two air conditioning systems are also the same as or similar to each other, that is, the pre-constructed self-encoders obtained by training according to the historical target parameters of the two air conditioning systems are also the same as or similar to each other, that is, the accurate pre-constructed self-encoders can be obtained by using the above method.
In some embodiments, the historical target parameter is a target parameter of the air conditioning system within a preset time period. Wherein the preset time duration is generally in the unit of months. Since the air conditioning system usually includes the target parameters under each operating condition after operating for 12 months, the preset time period is preferably 12 months.
And step S33, acquiring target parameters, wherein the target parameters comprise the operation parameters of the air conditioning system.
And step S34, inputting the target parameters into the pre-constructed self-encoder to obtain the output value of the pre-constructed self-encoder.
And step S35, determining whether the air conditioning system is abnormal or not according to the target parameter, the output value and the preset alarm threshold range.
And step S36, if the updating condition is met, updating the pre-constructed self-encoder according to the acquired target parameters.
In the embodiment of the application, the structural parameters of the self-encoder are determined according to the complexity of the air conditioning system, that is, the obtained pre-constructed self-encoder is guaranteed to be matched with the complexity of the air conditioning system, so that the more accurate pre-constructed self-encoder can be obtained after the self-encoder constructed based on the structural parameters is trained according to the historical target parameters. For example, if the air conditioning system a is more complex than the air conditioning system B, the number of the structural parameters corresponding to the air conditioning system a will be greater than the number of the structural parameters corresponding to the air conditioning system B. Because the number of the structural parameters corresponding to the air-conditioning system A is more than that of the structural parameters corresponding to the air-conditioning system B, even if the air-conditioning system A is more complex than the air-conditioning system B, the pre-constructed self-coding corresponding to the air-conditioning system A can accurately reflect the characteristics of the target parameters of the air-conditioning system A.
In some embodiments, before determining the structural parameters, the complexity of the air conditioning system is determined, that is, before the step S31, the method includes:
and A1, determining the number of the internal machines and the external machines of the air conditioning system.
And A2, determining the complexity of the air conditioning system according to the determined number of the internal machines and the external machines.
In the above-mentioned a1 and a2, the number of the internal machines and the external machines includes the number of the internal machines and the number of the external machines, and the external machines may be chiller units. Considering that the number of different chillers in the chiller unit is different, the complexity of the corresponding air conditioning system is different (the greater the number of chillers is, the higher the complexity of the corresponding air conditioning system is), and the number of different internal machines in the multi-split air conditioner is different, the complexity of the corresponding air conditioning system is also different (the greater the number of internal machines is, the higher the complexity of the corresponding air conditioning system is), therefore, the complexity of the air conditioning system can be determined according to whether the air conditioning system has chillers and internal machines, and the number of existing chillers and the number of internal machines. That is, according to the above manner, the complexity of the air conditioning system can be accurately determined, and then the accuracy of the subsequently obtained pre-constructed self-encoder and the obtained alarm threshold can be improved.
In some embodiments, before determining whether there is an abnormality in the air conditioning system, it is necessary to determine a preset alarm threshold range, that is, before step S35, the method includes:
and B1, calculating the mean square error of the historical target parameters and the output values.
And B2, calculating the mean value and the standard deviation of the mean square error according to the data distribution of the mean square error, and determining the preset alarm threshold range according to the mean value and the standard deviation.
In the above B1 and B2, the above mean square error MSE can be determined according to the following formula:
Figure BDA0003190150090000071
xias a historical target parameter, yiIs xiAnd (4) corresponding output values, wherein n is the number of the historical target parameters. In this embodiment, the upper and lower limits of the alarm threshold range may be determined following sigma principles. In some embodiments, assuming that the mean is expressed in μ and the standard deviation is expressed in σ, consider that the 1sigma principle is: the probability of the numerical distribution in (μ - σ, μ + σ) is 0.6526, the 2sigma principle is: the probability of the numerical distribution in (μ -2 σ, μ +2 σ) is 0.9544, while the 3sigma principle is: the probability of the value distribution in (μ -3 σ, μ +3 σ) is 0.9974, and therefore, the upper and lower limits of the alarm threshold range can be determined following the 3sigma principle to improve the probability of the target parameter distribution in the alarm threshold range. FIG. 4 shows a schematic diagram of the relationship between a target parameter, an upper limit of an alarm threshold range, and a lower limit of the alarm threshold range. In fig. 4, some target parameters exceed the upper limit of the alarm threshold range, and it is determined that there is an abnormality in the air conditioning system.
Example three:
fig. 5 is a flowchart illustrating a third anomaly detection method provided in the embodiment of the present application. In this embodiment, step S51 is the same as step S11 of the first embodiment, and step S55 is the same as step S14 of the first embodiment, which is not repeated herein. In addition, the present embodiment mainly adds a step of eliminating the influence of the environmental factors on the target parameters on the basis of the first embodiment, which is detailed as follows:
and step S51, acquiring target parameters, wherein the target parameters comprise the operation parameters of the air conditioning system.
And step S52, eliminating the influence of the environmental factors on the target parameters.
In this embodiment, considering that the same air conditioning system operates in different regions under different operating conditions, in order to improve the universality of the subsequently obtained abnormal determination result, the influence of the environmental factors on the obtained target parameters needs to be eliminated first.
For example, assuming the same air conditioning system, which operates in the south of the sea and in the northeast, respectively, the indoor temperatures of the two areas are cooled to the same temperature, respectively. Because the outdoor working conditions are different, the unit performance is also different.
In this embodiment, after the influence of the environmental factor is eliminated from the target parameter, a parameter unrelated to the environmental factor is obtained.
And step S53, inputting the target parameters after eliminating the influence into a pre-constructed self-encoder to obtain the output value of the pre-constructed self-encoder.
The pre-constructed self-encoder can be constructed according to the construction method of the self-encoder provided in the second embodiment, and details are not described here.
In this embodiment, since the input pre-constructed self-encoder is a parameter unrelated to the environmental factor, the output value output from the pre-constructed self-encoder reduces the influence of the operating condition, and thus, when the abnormality of the air conditioning system is subsequently determined according to the target parameter, the output value, and the like, which are not influenced by the environmental factor, a more accurate abnormality determination result is obtained.
And step S54, determining whether the air conditioning system is abnormal according to the target parameter after influence elimination, the output value and a preset alarm threshold range.
And step S55, if the updating condition is met, updating the pre-constructed self-encoder according to the acquired target parameters.
In the embodiment of the application, the influence of the environmental factors is eliminated on the obtained target parameters, and the target parameters which are not influenced by the environmental factors are used as the input of the pre-constructed self-encoder, so that the output value which is not influenced by the environmental factors can be obtained, and a more accurate abnormity judgment result can be obtained when the abnormity judgment is carried out on the air conditioning system according to the target parameters and the output value which are not influenced by the environmental factors and the preset alarm threshold value.
In some embodiments, the number of target parameters of this embodiment is greater than 1, and the step S52 includes:
and C1, selecting parameters of similar types from the target parameters, and determining a corresponding elimination processing mode according to the type of the selected parameters.
In this embodiment, the parameters of similar types do not refer to the parameters of the same type. For example, when two target parameters are both temperature data (e.g., one target parameter is a condensing temperature and the other target parameter is an evaporating temperature, i.e., the two target parameters are not both condensing temperatures and are not both evaporating temperatures), the two target parameters are similar types of parameters; when both target parameters are pressure data (e.g., one target parameter is exhaust pressure and the other target parameter is suction pressure, i.e., both target parameters are neither exhaust pressure nor suction pressure), the two target parameters are similar types of parameters. That is, when the units of two target parameters are the same, the types to which the two target parameters belong are similar types.
After classifying various similar type parameters from the target parameters, determining elimination processing modes corresponding to the different similar type parameters. That is, in the present embodiment, different similar types of parameters may correspond to different cancellation processing manners.
And C2, eliminating the influence of the environmental factors on the selected parameters according to the determined elimination processing mode.
In this embodiment, the similar type of parameters are processed in an elimination processing manner corresponding to the similar type of parameters, so that parameters that are not affected by environmental factors are obtained. For example, when two target parameters are respectively the exhaust pressure and the suction pressure, the elimination processing mode corresponding to the two target parameters is "division", that is, the exhaust pressure/the suction pressure, the obtained pressure ratio is a parameter for eliminating the influence of the environmental factor, and the influence of the environmental factor for eliminating the target parameters is equivalent to the working condition elimination processing. When the two target parameters are respectively the condensation temperature and the evaporation temperature, the elimination processing mode corresponding to the two target parameters is subtraction, namely the condensation temperature and the evaporation temperature, and the obtained temperature difference is the parameter for eliminating the influence of the environmental factors.
In the above-mentioned C1 and C2, since the selected target parameters are parameters with similar types, and the target parameters with similar types have relevance, that is, the target parameters with similar types have comparability, a more accurate elimination processing method can be determined according to the type to which the selected target parameters belong, and further, more accurate parameters that are not affected by environmental factors can be obtained.
In some embodiments, the target parameters of the embodiments of the present application further include an outdoor temperature, and when the two target parameters are a condenser outlet temperature (not a condensing temperature) and an outdoor temperature, respectively, the elimination processing manners corresponding to the two target parameters are subtraction, that is, the condenser outlet temperature — the outdoor temperature, and the obtained temperature difference is a parameter for eliminating the influence of the environmental factor. Because the target parameters also comprise the outdoor temperature, namely the target parameters not only comprise the operation parameters of the air conditioning system, but also have certain influence on the operation of the air conditioning system, after the influence of the environmental factors on the outdoor temperature and the outlet temperature of the condenser is eliminated, the parameters which are not influenced by the environmental factors can be obtained more comprehensively, and the accuracy of the subsequently obtained abnormal judgment result can be further improved.
In some embodiments, the step S55 (or step S14 or step S36) includes:
and if the updating condition is met, training the pre-constructed self-encoder according to the obtained target parameters to obtain the updated pre-constructed self-encoder.
In this embodiment, when it is determined that the update condition is satisfied, the obtained pre-constructed self-encoder is retrained in combination with the recently acquired target parameter. For example, if the update condition is that the set update time is reached, the pre-constructed self-encoder is updated with the target parameter in a period from the last update time to the current update time. Namely, the parameters in the pre-constructed self-encoder are updated by retraining the pre-constructed self-encoder, so as to obtain the updated self-encoder. The updated self-encoder is used as a new pre-constructed self-encoder, that is, after a new target parameter is subsequently acquired, a new pre-constructed self-encoder is input.
After the target parameters of the air conditioning system are obtained, the target parameters can be adopted to perform online updating on the deployed pre-constructed self-encoder, and the target parameters of the air conditioning system can better reflect the actual situation of the air conditioning system, so that after the pre-constructed self-encoder is retrained in the mode, a more accurate self-encoder can be obtained, and a more accurate abnormity judgment result can be obtained.
In some embodiments, step S13 (or step S35 or step S54) includes:
d1, determining the difference between the target parameter and the output value.
In this embodiment, the difference between the target parameter and the output value may be obtained by subtracting or dividing, for example, subtracting the target parameter from the output value to obtain a corresponding difference.
In some embodiments, the difference is determined in the same manner as the upper limit value and/or the lower limit value of the preset alarm threshold range. For example, if the upper limit value and/or the lower limit value of the preset alarm threshold range is determined according to the average value of the difference between the historical target parameter and the output value corresponding to the historical target parameter, the determination method of the difference here also adopts a subtraction processing method. The determination mode of the difference is consistent with the determination mode of the upper limit value and/or the lower limit value of the preset alarm threshold value range, namely, the difference and the upper limit value and/or the lower limit value are compared at the same latitude, so that the accuracy of the subsequently obtained abnormal judgment result can be improved.
D2, if the difference is not in the preset alarm threshold range, judging that the air conditioning system is abnormal.
D3, if the difference is within the preset alarm threshold range, judging that the air conditioning system is not abnormal.
Specifically, if the determined difference is determined to be a value within a preset alarm threshold range, it indicates that the air conditioning system is normal, otherwise, it indicates that the air conditioning system is abnormal. For example, if the determined difference is 0.5, the preset alarm threshold range is (0,1), and since 0.5 is a value between 0 and 1, it is determined that the air conditioning system is normal.
In some embodiments, the D1, above, includes:
and calculating the Euclidean distance between the target parameter and the output value to obtain the corresponding difference.
The above euclidean distance is also referred to as a euclidean distance. Since the Euclidean distance can express the real distance between two points in the m-dimensional space, the difference between the target parameter and the output value is more accurately represented by adopting the Euclidean distance. Wherein, the Euclidean distance is calculated by adopting the following formula:
Figure BDA0003190150090000101
wherein O represents the Euclidean distance, xiAs a historical target parameter, yiIs xiCorresponding output value, n being the total number of target parameters.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.
Example four:
fig. 6 shows a schematic structural diagram of an abnormality detection device provided in the embodiment of the present application, corresponding to the abnormality detection method in the above embodiment, and only the relevant parts of the embodiment of the present application are shown for convenience of description.
The abnormality detection device 6 is applied to an air conditioning system, and includes: a target parameter obtaining module 61, an output value determining module 62, an abnormality determining module 63, and a self-encoder updating module 64. Wherein:
and the target parameter acquiring module 61 is used for acquiring target parameters, and the target parameters comprise operation parameters of the air conditioning system.
In this embodiment, when the air conditioning system is in operation, the operation parameters generated during the operation of the air conditioning system, such as the compressor frequency, the condensation temperature, the evaporation temperature, the exhaust pressure, the suction pressure, and the like, are obtained.
In some embodiments, the target parameter further includes an environmental parameter, such as, for example, including outdoor temperature.
In some embodiments, since the obtained target parameter is to be used for detecting whether an abnormality occurs in the air conditioning system, in order to reduce the amount of data to be processed subsequently, the target parameter may be obtained after an interval duration arrives, where the interval duration is in units of seconds. For example, assuming that the duration of the interval is 5 seconds, the corresponding target parameter is acquired every 5 seconds. The target parameters corresponding to all the moments are not acquired, so that the number of the acquired target parameters can be effectively reduced, and the resources of the air conditioning system can be effectively saved. In addition, because the unit of the interval duration is 'second', namely the interval duration is short, even if the target parameters at each moment are not acquired, whether the air conditioning system is abnormal or not can be timely judged according to the subsequently acquired target parameters.
And the output value determining module 62 is configured to input the target parameter into the pre-constructed self-encoder to obtain an output value of the pre-constructed self-encoder.
In this embodiment, the structure of the self-encoder is a symmetric structure, such as the encoder in the self-encoder and the decoder in the self-encoder are symmetric. The number of hidden layers corresponding to the encoder and the decoder can be determined according to actual conditions. In some scenarios, assuming that the number of hidden layers corresponding to the encoder and the decoder are both 1, the structure of the self-encoder of this embodiment may be as shown in fig. 2. In fig. 2, the encoder includes an input layer and a first hidden layer; and the decoder comprises a third hidden layer and an output layer. As can be seen in fig. 2, the encoder and decoder of the self-encoder are symmetric.
In the present embodiment, the acquired target parameter is used as an input of the self-encoder, and after the characteristics of the input target parameter are learned by the self-encoder, a corresponding numerical value (i.e., the output value) is output.
And the abnormity determining module 63 is used for determining whether the air conditioning system is abnormal according to the target parameter, the output value and a preset alarm threshold range.
In this embodiment, a certain operation is performed on the target parameter and the output value, for example, a subtraction operation is performed on the target parameter and the output value, or a division operation is performed on the target parameter and the output value, so as to obtain a corresponding calculation result. And comparing the calculation result with a preset alarm threshold range, if the calculation result is within the preset alarm threshold range, judging that the air conditioning system is abnormal, otherwise, if the calculation result is not within the preset alarm threshold range, judging that the air conditioning system is abnormal.
And the self-encoder updating module 64 is configured to update the pre-constructed self-encoder according to the acquired target parameter if the updating condition is met.
In this embodiment, the update condition may be set according to an actual situation, for example, the update condition may be any of the following: the set update time is up, any target parameter (equivalent to real-time update) is acquired, and an update instruction sent by a user is received. For example, if the update condition is that the set update time is reached, the pre-constructed self-encoder will be automatically updated according to the acquired target parameters when the set update time is reached.
In some embodiments, if the pre-constructed self-encoder is not updated by acquiring any of the target parameters (i.e., the pre-constructed self-encoder is not updated in real time), the target parameters for updating the pre-constructed self-encoder may be: target parameters obtained during a time period after the last update action was performed and before the current update action was performed. For example, assume that the point in time at which the last update action was performed is 6 months, 5 days, 12: 00, the time point for executing the current update action is 6 months, 10 days, 12: 00, 6 months, 5 days 12: month 01 to 6, day 10 11: the target parameters acquired during this time period 59 are used to update the pre-constructed self-encoder. All the obtained target parameters are used for updating the pre-constructed self-encoder, namely, samples are increased, so that the method is beneficial to obtaining more accurate updated self-encoder.
In the embodiment of the application, since the target parameter comprises the operation parameter of the air conditioning system, after the target parameter is input into the self-encoder, the output value of the self-encoder is also related to the operation parameter, and therefore, whether the air conditioning system is abnormal or not can be judged according to the output value, the preset alarm threshold range and the target parameter. Meanwhile, compared with a fault detection model based on PCA, the self-encoder can adapt to an air conditioning system with higher complexity, so that after the characteristics of the target parameters are learned through the self-encoder of the embodiment of the application, more accurate output values can be obtained, and further, when abnormality judgment is carried out according to the more accurate output values, more accurate abnormality judgment results can be obtained. In addition, the self-encoder of the embodiment of the application can also be used for updating on line by combining the acquired target parameters, so that the anomaly detection method provided by the embodiment of the application can be suitable for air conditioning systems under different working conditions and/or different units, namely, even if the anomaly detection is carried out on the air conditioning systems under different working conditions and/or different units, an accurate anomaly judgment result can be obtained.
In some embodiments, the abnormality detection device 6 further includes: a structural parameter determination module and a pre-constructed self-encoder determination module, wherein:
and the structural parameter determining module is used for determining structural parameters according to the complexity of the air conditioning system.
In this embodiment, the complexity of the air conditioning system is related to the structure of the air conditioner itself and/or the function of the air conditioner, for example, the complexity of the air conditioning system corresponding to the multi-split air conditioner is higher than that of the single-split air conditioner. For example, air conditioning systems for air conditioners with multiple functions have a higher complexity than air conditioners with a single function.
In this embodiment, the structural parameters are structural parameters of an auto-encoder. Since the more complex the air conditioning system, the more variables are input, the more the number of neuron nodes per layer is caused. That is, when the air conditioning system is more complex, the corresponding structural parameters are more, and conversely, the corresponding structural parameters are less.
And the pre-constructed self-encoder determining module is used for acquiring the historical target parameters, training the self-encoder constructed based on the structural parameters according to the historical target parameters and obtaining the pre-constructed self-encoder.
In some embodiments, the acquired historical target parameters are target parameters corresponding to air conditioning systems similar to or identical to the subsequent air conditioning system that needs to be determined whether an abnormality exists. Since the air conditioning system where the historical target parameters are located is the same as or similar to the air conditioning system where the subsequently obtained target parameters are located, the historical target parameters obtained by the two air conditioning systems are also the same as or similar to each other, that is, the pre-constructed self-encoders obtained by training according to the historical target parameters of the two air conditioning systems are also the same as or similar to each other, that is, the accurate pre-constructed self-encoders can be obtained by adopting the above method.
In some embodiments, the historical target parameter is a target parameter of the air conditioning system within a preset time period. Wherein the preset time duration is generally in the unit of months. Since the air conditioning system usually includes the target parameters under each operating condition after operating for 12 months, the preset time period is preferably 12 months.
In some embodiments, the abnormality detection device 6 further includes: the device comprises a unit number determining module and a complexity determining module. Wherein:
and the unit number determining module is used for determining the number of the internal and external units of the air conditioning system.
And the complexity determining module is used for determining the complexity of the air conditioning system according to the determined number of the internal machine and the external machine.
In this embodiment, the number of the internal machines and the external machines includes the number of the internal machines and the number of the external machines, and the external machines may be water chilling units. Considering that the number of different chillers in the chiller unit is different, the complexity of the corresponding air conditioning system is different (the greater the number of chillers is, the higher the complexity of the corresponding air conditioning system is), and the number of different internal machines in the multi-split air conditioner is different, the complexity of the corresponding air conditioning system is also different (the greater the number of internal machines is, the higher the complexity of the corresponding air conditioning system is), therefore, the complexity of the air conditioning system can be determined according to whether the air conditioning system has chillers and internal machines, and the number of existing chillers and the number of internal machines. That is, according to the above manner, the complexity of the air conditioning system can be accurately determined, and then the accuracy of the subsequently obtained pre-constructed self-encoder and the obtained alarm threshold can be improved.
In some embodiments, the abnormality detection device 6 further includes: the device comprises a mean square error calculation module and a preset alarm threshold range determination module. Wherein:
and the mean square error calculation module is used for calculating the mean square error between the historical target parameters and the output value.
And the preset alarm threshold range determining module is used for calculating the mean value and the standard deviation of the mean square error according to the data distribution of the mean square error and determining the preset alarm threshold range according to the mean value and the standard deviation.
In this embodiment, the mean square error MSE may be determined according to the following formula:
Figure BDA0003190150090000131
wherein x isiAs a historical target parameter, yiIs xiAnd (4) corresponding output values, wherein n is the number of the historical target parameters. In this embodiment, the upper and lower limits of the alarm threshold range may be determined following sigma principles. In some embodiments, assuming that the mean is expressed in μ and the standard deviation is expressed in σ, consider that the 1sigma principle is: the probability of the numerical distribution in (μ - σ, μ + σ) is 0.6526, the 2sigma principle is: the probability of the numerical distribution in (μ -2 σ, μ +2 σ) is 0.9544, while the 3sigma principle is: the probability of the value distribution in (μ -3 σ, μ +3 σ) is 0.9974, and therefore, the upper and lower limits of the alarm threshold range can be determined following the 3sigma principle to improve the probability of the target parameter distribution in the alarm threshold range.
In some embodiments, the abnormality detection device 6 further includes: and an influence elimination module. Wherein:
and the influence elimination module is used for eliminating the influence of the environmental factors on the target parameters.
In this embodiment, considering that the same air conditioning system operates in different regions under different operating conditions, in order to improve the universality of the subsequently obtained abnormal determination result, the influence of the environmental factors on the obtained target parameters needs to be eliminated first.
For example, assuming the same air conditioning system, which operates in the south of the sea and in the northeast, respectively, the indoor temperatures of the two areas are cooled to the same temperature, respectively. Because the outdoor working conditions are different, the unit performance is also different.
In this embodiment, after the influence of the environmental factor is eliminated from the target parameter, a parameter unrelated to the environmental factor is obtained.
In some embodiments, the number of target parameters is greater than 1, and the influence elimination module includes: an elimination processing mode determining unit and an influence eliminating unit of the environmental factors. Wherein:
and the elimination processing mode determining unit is used for selecting parameters of similar types from the target parameters and determining the corresponding elimination processing mode according to the type of the selected parameters.
Wherein similar types of parameters do not refer to parameters of exactly the same type. For example, when two target parameters are both temperature data (e.g., one target parameter is a condensing temperature and the other target parameter is an evaporating temperature, i.e., the two target parameters are not both condensing temperatures and are not both evaporating temperatures), the two target parameters are similar types of parameters; when both target parameters are pressure data (e.g., one target parameter is exhaust pressure and the other target parameter is suction pressure, i.e., both target parameters are neither exhaust pressure nor suction pressure), the two target parameters are similar types of parameters. That is, when the units of two target parameters are the same, the types to which the two target parameters belong are similar types.
After classifying various similar type parameters from the target parameters, determining elimination processing modes corresponding to the different similar type parameters. That is, in the present embodiment, different similar types of parameters may correspond to different cancellation processing manners.
And the influence elimination unit of the environmental factors is used for eliminating the influence of the environmental factors on the selected parameters according to the determined elimination processing mode.
Specifically, the parameters of the similar type are processed by adopting an elimination processing mode corresponding to the parameters of the similar type, so that the parameters which are not influenced by the environmental factors are obtained. For example, when two target parameters are respectively the exhaust pressure and the suction pressure, the elimination processing mode corresponding to the two target parameters is "division", that is, the exhaust pressure/the suction pressure, the obtained pressure ratio is a parameter for eliminating the influence of the environmental factor, and the influence of the environmental factor for eliminating the target parameters is equivalent to the working condition elimination processing. When the two target parameters are respectively the condensation temperature and the evaporation temperature, the elimination processing mode corresponding to the two target parameters is subtraction, namely the condensation temperature and the evaporation temperature, and the obtained temperature difference is the parameter for eliminating the influence of the environmental factors.
In this embodiment, since the selected target parameters are parameters with similar types, and the target parameters with similar types have relevance, that is, the target parameters with similar types have comparability, a more accurate elimination processing mode can be determined according to the type to which the selected target parameters belong, and thus, more accurate parameters which are not affected by environmental factors can be obtained.
In some embodiments, the target parameters of the embodiments of the present application further include an outdoor temperature, and when the two target parameters are a condenser outlet temperature (not a condensing temperature) and an outdoor temperature, respectively, the elimination processing manners corresponding to the two target parameters are subtraction, that is, the condenser outlet temperature — the outdoor temperature, and the obtained temperature difference is a parameter for eliminating the influence of the environmental factor. Because the target parameters also comprise the outdoor temperature, namely the target parameters not only comprise the operation parameters of the air conditioning system, but also have certain influence on the operation of the air conditioning system, after the influence of the environmental factors on the outdoor temperature and the outlet temperature of the condenser is eliminated, the parameters which are not influenced by the environmental factors can be obtained more comprehensively, and the accuracy of the subsequently obtained abnormal judgment result can be further improved.
In some embodiments, the self-encoder update module 64 is specifically configured to:
and if the updating condition is met, training the pre-constructed self-encoder according to the obtained target parameters to obtain the updated pre-constructed self-encoder.
In this embodiment, when it is determined that the update condition is satisfied, the obtained pre-constructed self-encoder is retrained in combination with the recently acquired target parameter. For example, if the update condition is that the set update time is reached, the pre-constructed self-encoder is updated with the target parameter in a period from the last update time to the current update time. Namely, the parameters in the pre-constructed self-encoder are updated by retraining the pre-constructed self-encoder, so as to obtain the updated self-encoder. The updated self-encoder is used as a new pre-constructed self-encoder, that is, after a new target parameter is subsequently acquired, a new pre-constructed self-encoder is input.
After the target parameters of the air conditioning system are obtained, the target parameters can be adopted to perform online updating on the deployed pre-constructed self-encoder, and the target parameters of the air conditioning system can better reflect the actual situation of the air conditioning system, so that after the pre-constructed self-encoder is retrained in the mode, a more accurate self-encoder can be obtained, and a more accurate abnormity judgment result can be obtained.
In some embodiments, the anomaly determination module 63 includes: a difference determination unit, an abnormality presence determination unit, and an abnormality absence determination unit. Wherein:
a difference determining unit for determining a difference between the target parameter and the output value.
The difference between the target parameter and the output value may be obtained by subtracting or dividing, for example, subtracting the target parameter from the output value to obtain a corresponding difference.
In some embodiments, the difference is determined in the same manner as the upper limit value and/or the lower limit value of the preset alarm threshold range. For example, if the upper limit value and/or the lower limit value of the preset alarm threshold range is determined according to the average value of the difference between the historical target parameter and the output value corresponding to the historical target parameter, the determination method of the difference here also adopts a subtraction processing method. The determination mode of the difference is consistent with the determination mode of the upper limit value and/or the lower limit value of the preset alarm threshold value range, namely, the difference and the upper limit value and/or the lower limit value are compared at the same latitude, so that the accuracy of the subsequently obtained abnormal judgment result can be improved.
And the abnormity judging unit is used for judging that the air conditioning system is abnormal if the difference is not within the preset alarm threshold range.
And the non-existence abnormity determining unit is used for determining that the air conditioning system is not abnormal if the difference is within the preset alarm threshold range.
Specifically, if the determined difference is determined to be a value within a preset alarm threshold range, it indicates that the air conditioning system is normal, otherwise, it indicates that the air conditioning system is abnormal.
In some embodiments, the difference determining unit is specifically configured to:
and calculating the Euclidean distance between the target parameter and the output value to obtain the corresponding difference.
The above euclidean distance is also referred to as a euclidean distance. Since the Euclidean distance can express the real distance between two points in the m-dimensional space, the difference between the target parameter and the output value is more accurately represented by adopting the Euclidean distance. Wherein, the Euclidean distance is calculated by adopting the following formula:
Figure BDA0003190150090000161
wherein O represents the Euclidean distance, xiAs a historical target parameter, yiIs xiCorresponding output value, n being the total number of target parameters.
It should be noted that, for the information interaction, execution process, and other contents between the above-mentioned devices/units, the specific functions and technical effects thereof are based on the same concept as those of the embodiment of the method of the present application, and specific reference may be made to the part of the embodiment of the method, which is not described herein again.
Example five:
fig. 7 is a schematic structural diagram of an air conditioning system according to an embodiment of the present application. As shown in fig. 7, the air conditioning system 7 of this embodiment includes: at least one processor 70 (only one processor is shown in fig. 7), a memory 71, and a computer program 72 stored in the memory 71 and executable on the at least one processor 70, the steps of any of the above-described method embodiments being implemented when the computer program 72 is executed by the processor 70.
The air conditioning system 7 may be a desktop computer, a notebook, a palm computer, a cloud server, or other computing device. The air conditioning system 7 may include, but is not limited to, a processor 70, and a memory 71. Those skilled in the art will appreciate that fig. 7 is merely an example of air conditioning system 7, and does not constitute a limitation of air conditioning system 7, and may include more or less components than those shown, or some components in combination, or different components, and in one scenario may also include input output devices, network access devices, and the like.
The Processor 70 may be a Central Processing Unit (CPU), and the Processor 70 may be other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The storage 71 may in some embodiments be an internal storage unit of the air conditioning system 7, such as a hard disk or a memory of the air conditioning system 7. The memory 71 may also be an external storage device of the air conditioning system 7 in other embodiments, such as a plug-in hard disk provided on the air conditioning system 7, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like. Further, the memory 71 may also include both an internal storage unit of the air conditioning system 7 and an external storage device. The memory 71 is used for storing an operating system, an application program, a BootLoader (BootLoader), data, and other programs, such as program codes of a computer program. The memory 71 may also be used to temporarily store data that has been output or is to be output.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions. Each functional unit and module in the embodiments may be integrated in one processing unit, or each unit may exist alone physically, or two or more units are integrated in one unit, and the integrated unit may be implemented in a form of hardware, or in a form of software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working processes of the units and modules in the system may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
An embodiment of the present application further provides a network device, where the network device includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, the processor implementing the steps of any of the various method embodiments described above when executing the computer program.
The embodiments of the present application further provide a computer-readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the computer program implements the steps in the above-mentioned method embodiments.
The embodiments of the present application provide a computer program product, which when running on an air conditioning system, enables the air conditioning system to implement the steps in the above method embodiments when executed.
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, all or part of the processes in the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium and can implement the steps of the embodiments of the methods described above when the computer program is executed by a processor. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer readable medium may include at least: any entity or device capable of carrying computer program code to a photographing apparatus/terminal apparatus, a recording medium, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), an electrical carrier signal, a telecommunications signal, and a software distribution medium. Such as a usb-disk, a removable hard disk, a magnetic or optical disk, etc. In certain jurisdictions, computer-readable media may not be an electrical carrier signal or a telecommunications signal in accordance with legislative and patent practice.
In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and reference may be made to the related descriptions of other embodiments for parts that are not described or illustrated in a certain embodiment.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus/network device and method may be implemented in other ways. For example, the above-described apparatus/network device embodiments are merely illustrative, and for example, the division of the modules or units is only one logical division, and there may be other divisions when actually implementing, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not implemented. 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, devices or units, and may be in an electrical, mechanical 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 network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present application and are intended to be included within the scope of the present application.

Claims (12)

1. An abnormality detection method applied to an air conditioning system, comprising:
acquiring target parameters, wherein the target parameters comprise operation parameters of the air conditioning system;
inputting the target parameter into a pre-constructed self-encoder to obtain an output value of the pre-constructed self-encoder;
determining whether the air conditioning system is abnormal or not according to the target parameter, the output value and a preset alarm threshold range;
and if the updating condition is met, updating the pre-constructed self-encoder according to the acquired target parameter.
2. The abnormality detection method according to claim 1, characterized by, before said acquiring target parameters, comprising:
determining structural parameters according to the complexity of the air conditioning system;
and acquiring historical target parameters, and training a self-encoder constructed based on the structural parameters according to the historical target parameters to obtain the pre-constructed self-encoder.
3. The anomaly detection method according to claim 2, characterized in that, before said determining structural parameters according to the complexity of said air-conditioning system, it comprises:
determining the number of the internal machines and the external machines of the air conditioning system;
and determining the complexity of the air conditioning system according to the determined number of the internal machines and the external machines.
4. The abnormality detection method according to claim 2, wherein before said determining whether there is an abnormality of said air conditioning system based on said target parameter, said output value, and a preset alarm threshold range, comprising:
calculating the mean square error of the historical target parameter and the output value;
and calculating the average value and the standard deviation of the mean square error according to the data distribution of the mean square error, and determining the preset alarm threshold range according to the average value and the standard deviation.
5. The anomaly detection method according to claim 1, wherein said inputting said target parameter into a pre-constructed self-encoder comprises:
and eliminating the influence of environmental factors on the target parameters.
6. The abnormality detection method according to claim 5, wherein the number of said target parameters is greater than 1, and said eliminating the influence of the environmental factors on said target parameters includes:
selecting parameters of similar types from the target parameters, and determining a corresponding elimination processing mode according to the type of the selected parameters;
and eliminating the influence of the environmental factors on the selected parameters according to the determined elimination processing mode.
7. The anomaly detection method according to any one of claims 1 to 6, wherein said updating said pre-constructed self-encoder according to the acquired target parameter if an update condition is satisfied comprises:
and if the updating condition is met, training the pre-constructed self-encoder according to the obtained target parameters to obtain an updated pre-constructed self-encoder.
8. The abnormality detection method according to claim 7, wherein said determining whether there is an abnormality in said air conditioning system based on said target parameter, said output value, and a preset alarm threshold range includes:
determining a difference between the target parameter and the output value;
if the difference is not within the preset alarm threshold range, judging that the air conditioning system is abnormal;
and if the difference is within the preset alarm threshold range, judging that the air conditioning system is not abnormal.
9. The anomaly detection method of claim 8, said determining a difference in said target parameter and said output value comprising:
and calculating the Euclidean distance between the target parameter and the output value to obtain the corresponding difference.
10. An abnormality detection device, which is applied to an air conditioning system, includes:
the target parameter acquisition module is used for acquiring target parameters, and the target parameters comprise operation parameters of the air conditioning system;
the output value determining module is used for inputting the target parameter into a pre-constructed self-encoder to obtain the output value of the pre-constructed self-encoder;
the abnormity judging module is used for determining whether the air conditioning system is abnormal or not according to the target parameter, the output value and a preset alarm threshold range;
and the self-encoder updating module is used for updating the pre-constructed self-encoder according to the acquired target parameters if the updating conditions are met.
11. An air conditioning system comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the method according to any one of claims 1 to 9 when executing the computer program.
12. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1 to 9.
CN202110875546.1A 2021-07-30 2021-07-30 Abnormity detection method and device and air conditioning system Pending CN113587362A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110875546.1A CN113587362A (en) 2021-07-30 2021-07-30 Abnormity detection method and device and air conditioning system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110875546.1A CN113587362A (en) 2021-07-30 2021-07-30 Abnormity detection method and device and air conditioning system

Publications (1)

Publication Number Publication Date
CN113587362A true CN113587362A (en) 2021-11-02

Family

ID=78253082

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110875546.1A Pending CN113587362A (en) 2021-07-30 2021-07-30 Abnormity detection method and device and air conditioning system

Country Status (1)

Country Link
CN (1) CN113587362A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116754016A (en) * 2023-06-25 2023-09-15 北京华控智加科技有限公司 Fault detection method and device, electronic equipment and storage medium

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
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
CN109931678A (en) * 2019-03-13 2019-06-25 中国计量大学 Air-conditioning fault diagnosis method based on deep learning LSTM
CN111767183A (en) * 2020-09-01 2020-10-13 深圳创新奇智科技有限公司 Equipment abnormality detection method and device, electronic equipment and storage medium
CN111981326A (en) * 2020-08-12 2020-11-24 罗普特科技集团股份有限公司 Pipe gallery water leakage detection method and device
KR20210078256A (en) * 2019-12-18 2021-06-28 엘지전자 주식회사 Fault diagnosis device and method for optimizing fault diagnosis model

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
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
CN109931678A (en) * 2019-03-13 2019-06-25 中国计量大学 Air-conditioning fault diagnosis method based on deep learning LSTM
KR20210078256A (en) * 2019-12-18 2021-06-28 엘지전자 주식회사 Fault diagnosis device and method for optimizing fault diagnosis model
CN111981326A (en) * 2020-08-12 2020-11-24 罗普特科技集团股份有限公司 Pipe gallery water leakage detection method and device
CN111767183A (en) * 2020-09-01 2020-10-13 深圳创新奇智科技有限公司 Equipment abnormality detection method and device, electronic equipment and storage medium

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
刘芬: "《数据挖掘中的核心技术研究》", 31 March 2019, 地质出版社 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116754016A (en) * 2023-06-25 2023-09-15 北京华控智加科技有限公司 Fault detection method and device, electronic equipment and storage medium

Similar Documents

Publication Publication Date Title
Li et al. Anomaly detection with generative adversarial networks for multivariate time series
US10585774B2 (en) Detection of misbehaving components for large scale distributed systems
CN112188531B (en) Abnormality detection method, abnormality detection device, electronic apparatus, and computer storage medium
CN112150311B (en) Method, device, computer equipment and storage medium for detecting abnormal energy consumption state
JP6871877B2 (en) Information processing equipment, information processing methods and computer programs
CA3132346A1 (en) User abnormal behavior recognition method and device and computer readable storage medium
WO2014145977A1 (en) System and methods for automated plant asset failure detection
CN111538311B (en) Flexible multi-state self-adaptive early warning method and device for mechanical equipment based on data mining
CN110757510B (en) Method and system for predicting remaining life of robot
CN115378744B (en) Network security test evaluation system and method
CN111879349A (en) Sensor data deviation self-adaptive correction method
CN115936262B (en) Yield prediction method, system and medium based on big data environment interference
CN113587362A (en) Abnormity detection method and device and air conditioning system
CN111626360A (en) Method, device, equipment and storage medium for detecting fault type of boiler
JP7331369B2 (en) Abnormal Sound Additional Learning Method, Data Additional Learning Method, Abnormality Degree Calculating Device, Index Value Calculating Device, and Program
CN113111585A (en) Intelligent cabinet fault prediction method and system and intelligent cabinet
CN115066693A (en) Operation state classification system and operation state classification method
CN116723157A (en) Terminal behavior detection model construction method, device, equipment and storage medium
CN113048012B (en) Wind turbine generator yaw angle identification method and device based on Gaussian mixture model
CN114416467A (en) Anomaly detection method and device
JP7371694B2 (en) Deterioration detection method, deterioration detection program, and information processing device
CN111625525B (en) Environment data repairing/filling method and system
CN111221320B (en) Data interpretation method and device, terminal and computer readable medium
CN114358395A (en) Attendance checking prediction method and device
CN114172708A (en) Method for identifying network flow abnormity

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
RJ01 Rejection of invention patent application after publication

Application publication date: 20211102

RJ01 Rejection of invention patent application after publication