CN118296305A - Method and system for estimating heat received by refrigerant in refrigerator - Google Patents
Method and system for estimating heat received by refrigerant in refrigerator Download PDFInfo
- Publication number
- CN118296305A CN118296305A CN202410666283.7A CN202410666283A CN118296305A CN 118296305 A CN118296305 A CN 118296305A CN 202410666283 A CN202410666283 A CN 202410666283A CN 118296305 A CN118296305 A CN 118296305A
- Authority
- CN
- China
- Prior art keywords
- heat absorption
- refrigerant
- estimated value
- data
- absorption capacity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000003507 refrigerant Substances 0.000 title claims abstract description 92
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000010521 absorption reaction Methods 0.000 claims abstract description 175
- 238000007405 data analysis Methods 0.000 claims abstract description 13
- 230000000704 physical effect Effects 0.000 claims abstract description 8
- 238000005057 refrigeration Methods 0.000 claims description 33
- 238000004364 calculation method Methods 0.000 claims description 32
- 238000013528 artificial neural network Methods 0.000 claims description 27
- 238000007781 pre-processing Methods 0.000 claims description 27
- 238000012937 correction Methods 0.000 claims description 22
- 238000005070 sampling Methods 0.000 claims description 18
- 230000005856 abnormality Effects 0.000 claims description 17
- 230000002159 abnormal effect Effects 0.000 claims description 15
- 238000012549 training Methods 0.000 claims description 15
- 210000002569 neuron Anatomy 0.000 claims description 14
- 238000004422 calculation algorithm Methods 0.000 claims description 12
- 238000010606 normalization Methods 0.000 claims description 9
- 230000004913 activation Effects 0.000 claims description 5
- 238000011217 control strategy Methods 0.000 claims description 5
- 238000003062 neural network model Methods 0.000 claims description 5
- 230000003247 decreasing effect Effects 0.000 claims description 4
- 238000013461 design Methods 0.000 claims description 4
- 238000004806 packaging method and process Methods 0.000 claims description 2
- 239000012530 fluid Substances 0.000 claims 1
- 238000012545 processing Methods 0.000 abstract description 10
- 238000005516 engineering process Methods 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 10
- 238000005265 energy consumption Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 238000002474 experimental method Methods 0.000 description 4
- 238000005259 measurement Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 239000000306 component Substances 0.000 description 3
- 238000001704 evaporation Methods 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 238000000053 physical method Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000007619 statistical method Methods 0.000 description 3
- 238000001816 cooling Methods 0.000 description 2
- 238000013079 data visualisation Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000012217 deletion Methods 0.000 description 2
- 230000037430 deletion Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 230000008020 evaporation Effects 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- BASFCYQUMIYNBI-UHFFFAOYSA-N platinum Chemical compound [Pt] BASFCYQUMIYNBI-UHFFFAOYSA-N 0.000 description 2
- 230000001105 regulatory effect Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000009123 feedback regulation Effects 0.000 description 1
- 230000017525 heat dissipation Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013450 outlier detection Methods 0.000 description 1
- 229910052697 platinum Inorganic materials 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Air Conditioning Control Device (AREA)
Abstract
The present invention relates to the field of data processing technology, and in particular, to a method and a system for estimating heat received by a refrigerant in a refrigerator. The content comprises: firstly, operation parameters of a refrigerating system and physical property parameters of a refrigerant are collected and preprocessed in real time to obtain preprocessed time series data; combining the preprocessed time series data, and calculating the heat absorption and release quantity of the refrigerant in the compressor, the condenser and the evaporator to obtain an initial estimated value of the heat absorption quantity; correcting the initial estimated value of the heat absorption capacity to obtain an estimated value of the heat absorption capacity; further, defining a normal working interval of the estimated value of the heat absorption quantity, and judging whether the estimated value of the heat absorption quantity is in the normal working interval or not; and finally, applying the estimated value of the heat absorption capacity to energy-saving control and big data analysis of the refrigerating system. The method solves the technical problems of low precision, poor real-time performance and high cost existing in the existing refrigerant heat absorption capacity estimation method, and realizes real-time and accurate estimation of the refrigerant heat absorption capacity.
Description
Technical Field
The present invention relates to the field of data processing technology, and in particular, to a method and a system for estimating heat received by a refrigerant in a refrigerator.
Background
The refrigerator is used as a common refrigeration device and is widely applied to industrial production and daily life. The core component of the refrigerator is a compressor, which is used for increasing the pressure of the refrigerant to make the refrigerant condense and release heat in the condenser, and then reducing the pressure through a throttling device and then absorbing heat in an evaporator for evaporation, so that refrigeration is realized. One of the core indicators of a refrigerator is the amount of heat absorbed by the refrigerant in the evaporator, which directly affects the efficiency and energy consumption of the refrigeration system. In the refrigeration process, the evaluation of the heat absorption capacity of the refrigerant in the evaporator is a key for ensuring the efficient and energy-saving operation of the refrigerator. Currently, engineers typically use thermodynamic calculations or physical measurements to estimate the amount of refrigerant heat absorption.
Thermodynamic calculations are based on thermodynamic properties of the refrigerant and state parameters of the system, and the thermodynamic model is used to estimate the heat absorption of the refrigerant. For example, the temperature and pressure of the refrigerant at the inlet and outlet of the evaporator can be measured, the specific enthalpy value of the corresponding state point can be obtained by looking up a table, and then the heat absorption capacity of the refrigerant can be calculated by combining the mass flow. The method has the advantages of simple principle and less equipment investment, but is influenced by factors such as measurement errors, deterioration of working media, aging of a system and the like under actual working conditions, and has lower estimation accuracy. The commonly used temperature sensor comprises a thermocouple, a thermal resistor, a bimetal thermometer and the like, wherein the platinum resistor thermometer is widely applied with high measurement precision and good stability; common pressure sensors include resistive strain type, capacitive type, piezoelectric type, etc., where resistive strain type pressure sensors are dominant in the industrial field with high sensitivity, wide frequency response, simple structure. Another common method is physical measurement, i.e. by installing heat meters on both sides of the evaporator, directly measuring the heat exchange between the refrigerant and the secondary side medium. The method has high measurement accuracy, but has the defects of complex equipment and high cost, and larger system errors can be introduced when the measuring points are arranged improperly.
The defects directly lead to the problems of the refrigeration system such as reduced operation efficiency, increased energy consumption, aggravated component damage and the like, and increase the production and maintenance cost of enterprises. In summary, the existing refrigerant heat absorption capacity estimation method generally has the problems of low precision, poor real-time performance, high cost and the like. On one hand, because the thermodynamic calculation method is based on a steady working condition, the dynamic characteristics of the refrigeration system are difficult to reflect in real time, and part of influencing factors are ignored, so that an estimated value deviates from the reality; on the other hand, although the physical measurement method can directly obtain the instantaneous value of the heat absorption quantity, the measurement equipment is complex, the sensor is easily interfered by the outside, and the stability and the reliability of long-term operation are difficult to ensure. Therefore, how to realize real-time and accurate estimation of the heat absorption capacity of the refrigerant is a problem to be solved in the refrigeration field at present.
Disclosure of Invention
Based on the background art, in order to solve the above problems, the present invention provides a method and a system for estimating the heat received by a refrigerant in a refrigerator, which specifically include the following technical schemes:
a system for estimating the amount of heat received by a refrigerant in a chiller, comprising:
The system comprises a data acquisition module, a data preprocessing module, a heat absorption capacity calculation module, a neural network correction module, an abnormality judgment module and an application service module;
The data acquisition module acquires the operation parameters of the refrigerating system and the physical parameters of the refrigerant in real time; packaging the operation parameters of the refrigeration system and the physical parameters of the refrigerant into data frames, and publishing the data frames to a theme subscribed by a data preprocessing module through a Kafka message queue middleware;
the data preprocessing module is used for preprocessing the operation parameters of the refrigerating system and the physical parameters of the refrigerant to obtain preprocessed time series data; the preprocessed time series data are published to a Kafka theme subscribed by the heat absorption capacity calculation module;
the heat absorption quantity calculation module is used for calculating the heat absorption quantity of the refrigerant in the compressor, the condenser and the evaporator on the basis of classical thermodynamic theory by combining the preprocessed time series data to obtain an initial estimated value of the heat absorption quantity; publishing the initial estimated value of the heat absorption capacity to a Kafka theme subscribed by the neural network correction module;
the neural network correction module corrects the initial estimated value of the heat absorption capacity by adopting a neural network model to obtain an estimated value of the heat absorption capacity; releasing the estimated value of the heat absorption capacity to a Kafka theme subscribed by the abnormality judgment module;
The abnormality judgment module is responsible for evaluating whether the estimated value of the heat absorption capacity is in a normal working interval; releasing the estimated value of the heat absorption falling in the normal working interval to an application service module, judging the estimated value of the heat absorption exceeding the normal working interval as abnormal, and releasing an alarm signal to the application service module;
the application service module is used for applying the estimated value of the heat absorption capacity to energy-saving control and big data analysis of the refrigerating system.
Preferably, the preprocessing in the data preprocessing module is implemented as follows:
Detecting abnormal values of the operation parameters of the refrigeration system at each sampling moment by adopting a3 sigma criterion, and filling the abnormal values; performing interpolation estimation on the operation parameters of the refrigeration system and the missing values in the physical parameters of the refrigerant; further adopting a maximum and minimum value normalization method to perform standardization treatment; and taking the time stamp of the temperature data as a reference, and carrying out linear interpolation on data except the temperature data in the operation parameters of the refrigerating system and the physical parameters of the refrigerant so as to align the time stamp with the temperature data.
Preferably, the heat absorption capacity calculation module calculates the heat absorption capacity of the refrigerant of the compressor, the condenser and the evaporator respectively through a steady-state working medium flow energy balance equation based on the preprocessed time series data, and takes the heat absorption capacity of the compressor, the heat absorption capacity of the condenser and the heat absorption capacity of the evaporator as initial estimated values of the heat absorption capacity.
Preferably, the neural network correction module corrects the initial estimated value of the heat absorption capacity by adopting a three-layer BP neural network, the BP neural network is trained by adopting a gradient descent algorithm, and the estimated value of the heat absorption capacity of the whole machine is predicted by using the trained BP neural network based on the heat absorption capacity of the compressor, the heat release capacity of the condenser and the heat absorption capacity of the evaporator.
Preferably, the network structure of the BP neural network is as follows:
Input layer: 3 neurons for respectively receiving heat absorption capacity of the compressor, heat release capacity of the condenser and heat absorption capacity of the evaporator;
Hidden layer: 10 neurons, activation function is ReLU;
Output layer: and 1 neuron, outputting the corrected estimated heat absorption quantity value of the whole machine.
Preferably, the abnormality judgment module determines a normal working interval of the estimated value of the heat absorption capacity according to the design specification and the working condition constraint of the refrigerating system.
Preferably, in the aspect of energy saving control in the application service module, the adjustment amounts of the compressor frequency and the expansion valve opening degree are calculated by a PID control algorithm based on the estimated value of the heat absorption amount in combination with the control target of the refrigeration system.
Preferably, when the estimated value of the heat absorption amount is smaller than the target heat absorption amount, the frequency of the compressor should be appropriately increased or the opening degree of the expansion valve should be increased; when the estimated value of the heat absorption amount is larger than the target heat absorption amount, the compressor frequency should be appropriately decreased or the expansion valve opening degree should be decreased.
Preferably, in terms of big data analysis, the application service module uploads the estimated value of the heat absorption capacity, the frequency of the compressor, the opening data of the expansion valve and the alarm signal of each sampling period to the cloud platform, and the cloud platform stores and manages the data by adopting the time sequence database after receiving the data.
A method for estimating the heat of a refrigerant in a refrigerator, for implementing the above system for estimating the heat of a refrigerant in a refrigerator, comprising the steps of:
Firstly, operation parameters of a refrigerating system and physical property parameters of a refrigerant are collected and preprocessed in real time to obtain preprocessed time series data;
further, combining the preprocessed time series data, calculating the heat absorption and release quantity of the refrigerant in the compressor, the condenser and the evaporator to obtain an initial estimated value of the heat absorption quantity; correcting the initial estimated value of the heat absorption capacity to obtain an estimated value of the heat absorption capacity;
further, defining a normal working interval of the estimated value of the heat absorption quantity, and judging whether the estimated value of the heat absorption quantity is in the normal working interval or not;
And finally, applying the estimated value of the heat absorption capacity to energy-saving control and big data analysis of the refrigerating system, and optimizing a control strategy of the refrigerating system.
The technical scheme of the invention has the beneficial effects that:
1. The invention solves the problems of inaccurate and non-real-time estimation of the heat absorption quantity of the refrigerant caused by the lag of the measuring means and complex and changeable working conditions of the traditional method through the cooperative work of each functional module, thereby realizing the real-time and accurate estimation of the heat absorption quantity of the refrigerant and providing reliable basis for the energy-saving control of a refrigerating system.
2. In the data acquisition and preprocessing links, the invention adopts a plurality of sensors to acquire state parameters such as temperature, pressure, flow and the like of the refrigerating system in real time, combines a refrigerant physical property database to acquire thermodynamic property parameters, and overcomes the problems of original data such as deletion, abnormality, asynchronism, inconsistent dimension and the like by processing such as data cleaning, deletion value interpolation, abnormal value correction, time stamp alignment, normalization and the like, thereby outputting high-quality standardized time series data and laying a solid foundation for subsequent calculation.
3. In the heat absorption calculation link, the invention estimates the initial value of the heat absorption of the refrigerant in the parts such as the compressor, the condenser, the evaporator and the like based on classical thermodynamic theory by utilizing the preprocessed data, overcomes the problem that the traditional method is difficult to model and describe complex working conditions, and further obtains the initial estimated value of the heat absorption, although the estimated value still has a certain error.
4. In order to further improve the estimation accuracy, the three-layer BP neural network model is built through a deep learning technology, and the problem of estimation deviation caused by simplifying assumptions of a classical physical model is solved by learning the mapping rule of the heat absorption capacity through offline training, so that the estimated value of the heat absorption capacity of the whole machine is more approximate to a true value.
5. In an application service link, the invention applies the estimated value of the heat absorption to PID feedback regulation of energy-saving control of the refrigeration system, overcomes the problem of poor dynamic response of a conventional control method, further realizes that the refrigeration system reduces energy consumption while meeting comfort level, and uploads the estimated value of the historical heat absorption to a cloud platform for big data analysis, overcomes the problem of low data utilization rate of the traditional refrigeration system, and further provides data support for energy-saving optimization and predictive maintenance of the refrigeration system.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention to achieve the preset purpose, the technical solutions of the embodiments of the present invention will be clearly and completely described below, and it is obvious that the described embodiments are only some embodiments, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The invention provides a method and a system for estimating the heat of a refrigerant in a refrigerator, which concretely comprise the following technical scheme:
a system for estimating the amount of heat received by a refrigerant in a chiller, comprising:
The system comprises a data acquisition module, a data preprocessing module, a heat absorption capacity calculation module, a neural network correction module, an abnormality judgment module and an application service module;
the data acquisition module is responsible for acquiring the operation parameters of the refrigeration system in real time and acquiring the physical parameters of the refrigerant by inquiring a thermodynamic property database of the refrigerant, so as to provide data support for subsequent calculation and analysis;
Specifically, the data acquisition module acquires the temperature reflecting the state of the refrigerant in real time at a sampling frequency of 1 second by installing a temperature sensor, a pressure sensor, a flowmeter and other instruments at a key position of the refrigeration system Pressure and forceMass flow rateEtc. operating parameters. The key positions of the refrigerating system are such as compressor inlet and outlet, condenser inlet and outlet, front and back of expansion valve, evaporator inlet and outlet, etc. Considering the difference of thermodynamic properties of the refrigerants with different types, the data acquisition module also needs to query the corresponding thermodynamic property databases of the refrigerants with the corresponding types according to the refrigerant types marked on the nameplate of the refrigeration system, such as R22, R134a and the like, so as to obtain the specific volume of the refrigerant with the type under different temperature and pressure conditions) Specific enthalpy%) Specific entropy%) And other physical parameters. These physical parameters may be obtained by consulting a refrigerant thermodynamic properties table or calling an API interface of commercial thermodynamic calculation software such as Refprop, coolProp.
The operation parameters and physical parameters acquired by the data acquisition module are packaged into data frames in a fixed data format such as JSON, and are released into the subject subscribed by the data preprocessing module through message queue middleware such as Kafka.
The data preprocessing module is responsible for preprocessing the acquired and queried original data such as outlier detection and correction, missing value filling and standardization processing to obtain preprocessed time sequence data, improve data quality and lay a foundation for subsequent calculation and modeling.
First, a data preprocessing module subscribes to a data frame published by a data acquisition module and receives various original data such as temperature, pressure, mass flow, specific volume, specific enthalpy, specific entropy and the like. Then, the data preprocessing module respectively processes various original data according to the data types as follows:
for temperature, pressure, mass flow data, the collected data may have outliers in consideration of sensor faults, signal interference, and the like. Therefore, the data preprocessing module adopts the 3 sigma criterion to detect the abnormal value of the data at each sampling moment, namely, if a certain data point deviates from the average value of the data in the sampling window by more than 3 times of standard deviation, the abnormal value is judged. The correction of the outlier can be achieved by various methods, such as a front-back value averaging method, hampel filtering method, kalman filtering method and the like, and an appropriate method needs to be selected according to the distribution characteristics and physical significance of the outlier. The invention selects a front-back value average method, namely, the data average value of two normal sampling points before and after an abnormal point is used for replacing the abnormal value.
For various original data, due to equipment failure, communication interruption and the like, the situation of data missing can occur at certain moments. To ensure the continuity of time series data, interpolation estimation of the missing values is required. The data preprocessing module adopts a linear interpolation method to realize missing value estimation, namely, the missing value is replaced by linear fitting values of data of two normal sampling points before and after the missing point. If the missing points appear at the beginning and end of the time series, then the nearest neighbor normal values are used for filling.
In order to improve the comparability of data and the stability of numerical calculation, dimensionless processing is required to be carried out on the data after the anomaly correction and the missing interpolation in consideration of the difference of the dimension and the magnitude of the data of different physical quantities. The data preprocessing module adopts a maximum and minimum normalization method to perform standardization processing, and linearly maps the data of each physical quantity to a [0,1] interval, wherein the formula is as follows:
，
Wherein, For the value after the normalization,For the value before normalization,AndThe minimum value and the maximum value of the physical quantity, respectively.
Finally, due to the different physical quantity data sources and processing links, the time stamp of the data may be misplaced. To ensure synchronization of data of each physical quantity in the time dimension, it is necessary to time-stamp data of other physical quantities with reference to the time stamp of a certain physical quantity. The data preprocessing module takes the time stamp of the temperature data as a reference, and performs linear interpolation on the data of other physical quantities so as to align the time stamp with the temperature data.
After the pretreatment is finished, standardized temperature, pressure, mass flow, specific volume, specific enthalpy and specific entropy time series data are output and are published to a Kafka theme subscribed by the heat absorption capacity calculation module.
The heat absorption quantity calculation module is used for carrying out preliminary calculation on the heat absorption quantity of the refrigerant in the components such as the compressor, the condenser, the evaporator and the like by combining the preprocessed time series data on the basis of classical thermodynamic theory.
First, the heat absorption capacity calculation module subscribes to the standardized time series data published by the data preprocessing module. These data include: inlet temperature of compressorInlet pressureOutlet temperatureOutlet pressure; Compressor mass flow; Condenser inlet temperatureInlet pressureOutlet temperatureOutlet pressure; Evaporator inlet temperatureInlet pressureOutlet temperatureOutlet pressure；
Meanwhile, the heat absorption capacity calculation module also needs to query the specific enthalpy of the corresponding state point from the refrigerant physical property database according to the temperature and the pressure of the state point；
By utilizing the state parameters, the heat absorption capacity calculation module calculates the heat absorption capacity of the refrigerant of the compressor, the condenser and the evaporator based on a steady-state working medium flow energy balance equation, wherein the formula is as follows:
heat absorption capacity of compressor ：
，
Wherein,AndSpecific enthalpy of refrigerant at the compressor inlet and outlet, respectively.
Heat release of condenser：
，
Wherein,AndSpecific enthalpy of refrigerant at the condenser inlet and outlet, respectively.
Heat absorption capacity of evaporator：
，
Wherein,AndSpecific enthalpy of refrigerant at the evaporator inlet and outlet, respectively.
Here, the heat absorption capacity of the compressorUnder ideal conditions, the total heat release amount of the condenser and the heat absorption amount of the evaporator are equal to each other, namely. However, under actual working conditions, the three components are not completely equal due to the influence of factors such as heat dissipation, throttling loss and the like of the pipeline. In order to simplify the calculation, the scheme does not consider the influence factors, and correction coefficients can be further introduced in the future to improve the calculation accuracy.
Output the heat absorption capacity calculation module、、The time series data is used as an initial estimated value of the heat absorption quantity and is published into a Kafka theme subscribed by the neural network correction module.
The neural network correction module adopts a deep learning technology to intelligently correct the initial estimated value of the heat absorption quantity, overcomes errors caused by simplifying assumptions of a classical thermodynamic model, and improves the accuracy of the heat absorption quantity estimation.
The neural network correction module is realized by adopting a three-layer BP neural network, and the network structure is as follows:
input layer: 3 neurons for respectively receiving heat absorption capacity of compressor Heat release of condenserHeat absorption capacity of evaporatorIs determined by the initial estimate of (a);
Hidden layer: 10 neurons, activation function is ReLU;
Output layer: 1 neuron, outputs the corrected estimated value of the heat absorption capacity of the whole machine 。
Wherein, the expression of the ReLU activation function is:
，
The loss function uses Mean Square Error (MSE) expressed as:
，
Wherein, In order to train the number of samples,Is the firstThe whole machine of each training sample is actually measured for heat absorption,Is the firstThe complete machine of each training sample predicts the heat absorption capacity.
The network training employs a gradient descent algorithm to minimize the loss function by continually updating neuron weights and biases through back propagation. And stopping training and storing network parameters when the change rate of the loss function is lower than a threshold set according to an expert experience method or the number of training rounds reaches an upper limit.
The neural network correction module needs to prepare a batch of training samples in advance. Training samples were obtained by designing refrigerant heat absorption test experiments, each sample containing four fields:、、 And The values of the first three fields are estimated by the heat absorption capacity calculation module, and the value of the last field is accurately measured by a heat metering device of a test experiment.
The neural network correction module subscribes to the release of the heat absorption capacity calculation module、、Data, using a trained neural network model, predicting the estimated value of the heat absorption capacity of the whole machineAnd publishing the topic to the Kafka topic subscribed by the abnormality judgment module.
The abnormality judgment module is responsible for evaluating whether the estimated value of the heat absorption quantity output by the neural network is in a normal working range or not, and is a safety valve of the refrigerating system.
Firstly, an abnormality judgment module needs to determine a normal working interval of an estimated value of heat absorption capacity according to design specification and working condition constraint of a refrigerating system. Generally, the heat absorption capacity under the rated working condition can be improvedAs a reference, multiply by a margin coefficient() Obtaining the upper and lower limits of the interval:
，
For example, take The normal working interval is。
The anomaly judgment module subscribes to the estimated value of the heat absorption capacity of the whole machine published by the neural network correction moduleAnd combine it withInterval comparison:
If it is If the estimated value falls within the interval, the estimated value of the heat absorption is considered to be normal and issuedTo an application service module; if it isAnd if the estimated value exceeds the interval, the estimated value of the heat absorption quantity is considered to be abnormal, and an alarm signal is issued to the application service module to prompt an engineer to check the cause of the abnormality.
The application service module is used for applying the estimated value of the heat absorption to the energy-saving control of the refrigerating system on one hand and uploading the estimated value of the historical heat absorption to the cloud platform for big data analysis on the other hand.
In the aspect of energy-saving control, an application service module subscribes to the estimated value of the heat absorption quantity released by the abnormality judgment module(Skipping if abnormality is found), calculating the frequency of the compressor by a PID control algorithm in combination with the control objective of the refrigeration systemOpening of expansion valveIs used for adjusting the adjustment quantity of the (a);
When (when) Less than the target heat absorptionWhen the flow rate of the refrigerant is smaller or the evaporation temperature is too low, the frequency of the compressor should be properly increasedOr increasing the opening of the expansion valve; When (when)Is greater than the target heat absorption capacityAt the same time, the evaporating temperature is too high, and the frequency of the compressor should be properly reducedOr reducing the opening of the expansion valve。
The discretized form of the PID control algorithm is:
，
Wherein, Is the firstThe control quantity output at each sampling instant,Is the firstThe difference between the target value and the measured value at each sampling instant, i.e，、、The coefficients of proportion, integral and differential are respectively determined by empirical setting methods such as Ziegler-Nichols; Is the first And estimating the heat absorption capacity of the whole machine at each sampling moment.
The application service module calculates the result、The regulating variable is sent to the executing mechanisms such as the frequency converter, the electronic expansion valve and the like through the field bus network such as RS485, CAN and the like, so that the feedback control of the refrigerating system is realized.
In terms of big data analysis, the application service module estimates the heat absorption capacity of each sampling periodCompressor frequencyOpening of expansion valveAnd uploading the data and the alarm signals to the cloud platform through the communication protocols of the Internet of things such as the MQTT. After the cloud platform receives the data, a time sequence database such as InfluxDB, openTSDB is adopted for storage and management, and data services such as data visualization, statistical analysis, anomaly detection, fault diagnosis and the like are provided.
For example by drawingThe time sequence curve of the cooling system is subjected to statistical analysis, so that the heat absorption level and the change rule of the cooling system under different working conditions can be mastered; analysis by data mining algorithmThe relevance with the system energy consumption of the refrigeration system can establish an energy consumption prediction model to guide energy saving optimization; by tracking analysisBy combining with a machine learning algorithm, potential faults of the refrigerating system can be found and positioned early, and predictive maintenance is realized.
Finally, the cloud platform opens the analysis result to the end user in the form of Web service, supports various access modes such as PC, mobile terminal and the like, is convenient for engineers and managers to know the running state of the refrigeration system anytime and anywhere, and optimizes the control strategy of the refrigeration system.
In addition to the above system for estimating the heat of refrigerant in a refrigerator, there is also provided a method for estimating the heat of refrigerant in a refrigerator, comprising the steps of:
s1, acquiring operation parameters of a refrigeration system and physical property parameters of a refrigerant in real time;
by installing temperature sensors, pressure sensors, flow meters and other instruments at key positions of the refrigeration system, such as a compressor inlet and outlet, a condenser inlet and outlet, front and back of an expansion valve, an evaporator inlet and outlet and the like, the operation parameters of temperature, pressure, mass flow and the like which reflect the state of the refrigerant are collected in real time. And according to the type of the refrigerant marked on the nameplate of the refrigeration system, inquiring a corresponding thermodynamic property database of the refrigerant to obtain physical parameters such as specific volume, specific enthalpy, specific entropy and the like of the type of refrigerant under different temperature and pressure conditions. These physical parameters may be obtained by consulting a refrigerant thermodynamic properties table or calling an API interface of commercial thermodynamic calculation software such as Refprop, coolProp.
S2, preprocessing the operation parameters of the refrigeration system and the physical property parameters of the refrigerant;
first, a data preprocessing module subscribes to a data frame published by a data acquisition module and receives various original data such as temperature, pressure, mass flow, specific volume, specific enthalpy, specific entropy and the like. Then, the data preprocessing module respectively processes various original data according to the data types as follows:
for temperature, pressure, mass flow data, use 3 And detecting the abnormal value of the data at each sampling time according to the criterion, and correcting the abnormal value by a front-back value average method.
Further, a linear interpolation method is adopted for the missing values in various original data to estimate the missing values. If the missing points appear at the beginning and end of the time series, then the nearest neighbor normal values are used for filling.
Taking the difference of the dimension and magnitude of the data of different physical quantities into consideration, carrying out standardization processing by adopting a maximum and minimum value normalization method, and linearly mapping the data of each physical quantity to a [0,1] interval, wherein the formula is as follows:
，
Wherein, For the value after the normalization,For the value before normalization,AndThe minimum value and the maximum value of the physical quantity, respectively.
Finally, due to the different physical quantity data sources and processing links, the time stamp of the data may be misplaced. And taking the time stamp of the temperature data as a reference, and linearly interpolating the data of other physical quantities to align the time stamp with the temperature data.
After the pretreatment is finished, standardized temperature, pressure, mass flow, specific volume, specific enthalpy and specific entropy time series data are output.
S3, combining the preprocessed time series data, and calculating the heat absorption and release quantity of the refrigerant in the compressor, the condenser and the evaporator to obtain an initial estimated value of the heat absorption quantity;
the normalized time series data includes: inlet temperature of compressor Inlet pressureOutlet temperatureOutlet pressure; Compressor mass flow; Condenser inlet temperatureInlet pressureOutlet temperatureOutlet pressure; Evaporator inlet temperatureInlet pressureOutlet temperatureOutlet pressure；
Meanwhile, the heat absorption capacity calculation module inquires the specific enthalpy of a corresponding state point from a refrigerant physical property database according to the temperature and the pressure of the state point；
By using the state parameters, the refrigerant heat absorption and release quantity of the compressor, the condenser and the evaporator is calculated based on a steady-state working medium flow energy balance equation, wherein the formula is as follows:
heat absorption capacity of compressor ：
，
Wherein,AndSpecific enthalpy of refrigerant at the compressor inlet and outlet, respectively.
Heat release of condenser：
，
Wherein,AndSpecific enthalpy of refrigerant at the condenser inlet and outlet, respectively.
Heat absorption capacity of evaporator：
，
Wherein,AndSpecific enthalpy of refrigerant at the evaporator inlet and outlet, respectively.
Output the heat absorption capacity calculation module、、As an initial estimate of the amount of heat absorption.
S4, correcting the initial estimated value of the heat absorption capacity to obtain an estimated value of the heat absorption capacity;
the correction of the initial estimated value of the heat absorption capacity is realized by adopting a three-layer BP neural network, and the network structure is as follows:
input layer: 3 neurons for respectively receiving heat absorption capacity of compressor Heat release of condenserHeat absorption capacity of evaporatorIs determined by the initial estimate of (a);
Hidden layer: 10 neurons, activation function is ReLU;
Output layer: 1 neuron, outputs the corrected estimated value of the heat absorption capacity of the whole machine 。
The loss function uses Mean Square Error (MSE) expressed as:
，
Wherein, In order to train the number of samples,Is the firstThe whole machine of each training sample is actually measured to absorb heat; Is the first The complete machine of each training sample predicts the estimated value of heat absorption.
The network training employs a gradient descent algorithm to minimize the loss function by continually updating neuron weights and biases through back propagation. And stopping training and storing network parameters when the change rate of the loss function is lower than a threshold set according to an expert experience method or the number of training rounds reaches an upper limit.
Training samples were obtained by designing refrigerant heat absorption test experiments, each sample containing four fields:、、 And The values of the first three fields are estimated by the heat absorption capacity calculation module, and the value of the last field is accurately measured by a heat metering device of a test experiment.
The neural network correction module subscribes to the release of the heat absorption capacity calculation module、、Data, using a trained neural network model, predicting the estimated value of the heat absorption capacity of the whole machine。
S5, defining a normal working interval of the estimated value of the heat absorption quantity, and judging whether the estimated value of the heat absorption quantity is in the normal working interval or not;
determining a normal working interval of the estimated value of the heat absorption capacity according to the design specification and the working condition constraint of the refrigerating system The specific formula is as follows:
，
Wherein, The heat absorption capacity is the heat absorption capacity under the rated working condition; As a margin coefficient of the degree of freedom, 。
Estimating the heat absorption quantity of the whole machineAnd (3) withThe intervals are compared:
If it is If the estimated value falls within the interval, the estimated value of the heat absorption quantity is considered to be normal; if it isIf the estimated value exceeds the interval, the estimated value of the heat absorption quantity is considered to be abnormal, and an alarm signal is required to be issued to prompt an engineer to check the cause of the abnormality.
S6, applying the estimated value of the heat absorption capacity to energy-saving control and big data analysis of the refrigerating system, and optimizing a control strategy of the refrigerating system.
In the aspect of energy-saving control, the estimated value based on the heat absorption capacity of the whole machine(Skipping if abnormality is found), calculating the frequency of the compressor by a PID control algorithm in combination with the control objective of the refrigeration systemOpening of expansion valveIs used for adjusting the adjustment quantity of the (a);
The discretized form of the PID control algorithm is:
，
Wherein, Is the firstOutputting control quantity of each sampling moment; Is the first The difference between the target value and the measured value at each sampling instant, i.e；Absorbing heat for a target; Is the first Estimating the heat absorption capacity of the whole machine at each sampling moment;、、 the ratio, integral and differential coefficients can be determined by Ziegler-Nichols and other empirical setting methods.
Will be、The regulating variable of the control system is sent to the actuating mechanisms such as a frequency converter, an electronic expansion valve and the like through a field bus network such as RS485, CAN and the like, so that the feedback control of the refrigerating system is realized.
In terms of big data analysis, the estimated value of the heat absorption capacity of each sampling periodCompressor frequencyOpening of expansion valveAnd uploading the data and the alarm signals to the cloud platform through the communication protocols of the Internet of things such as the MQTT. And after the cloud platform receives the data, the time sequence database is adopted for storage and management, and data services such as data visualization, statistical analysis, anomaly detection, fault diagnosis and the like are provided.
Finally, the cloud platform opens the big data analysis result to the end user in the form of Web service, supports various access modes such as PC, mobile terminal and the like, is convenient for engineers and managers to know the running state of the refrigeration system anytime and anywhere, and optimizes the control strategy of the refrigeration system.
In view of the foregoing, a method and system for estimating the heat received by a refrigerant in a refrigerator has been completed.
The sequence of the embodiments of the invention is merely for description and does not represent the advantages or disadvantages of the embodiments. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and the same or similar parts of each embodiment are referred to each other, and each embodiment mainly describes differences from other embodiments.
The above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention, and are intended to be included in the scope of the present invention.
Claims (10)
1. A system for estimating the amount of heat received by a refrigerant in a chiller, comprising:
The system comprises a data acquisition module, a data preprocessing module, a heat absorption capacity calculation module, a neural network correction module, an abnormality judgment module and an application service module;
The data acquisition module acquires the operation parameters of the refrigerating system and the physical parameters of the refrigerant in real time; packaging the operation parameters of the refrigeration system and the physical parameters of the refrigerant into data frames, and publishing the data frames to a theme subscribed by a data preprocessing module through a Kafka message queue middleware;
the data preprocessing module is used for preprocessing the operation parameters of the refrigerating system and the physical parameters of the refrigerant to obtain preprocessed time series data; the preprocessed time series data are published to a Kafka theme subscribed by the heat absorption capacity calculation module;
the heat absorption quantity calculation module is used for calculating the heat absorption quantity of the refrigerant in the compressor, the condenser and the evaporator on the basis of classical thermodynamic theory by combining the preprocessed time series data to obtain an initial estimated value of the heat absorption quantity; publishing the initial estimated value of the heat absorption capacity to a Kafka theme subscribed by the neural network correction module;
the neural network correction module corrects the initial estimated value of the heat absorption capacity by adopting a neural network model to obtain an estimated value of the heat absorption capacity; releasing the estimated value of the heat absorption capacity to a Kafka theme subscribed by the abnormality judgment module;
The abnormality judgment module is responsible for evaluating whether the estimated value of the heat absorption capacity is in a normal working interval; releasing the estimated value of the heat absorption falling in the normal working interval to an application service module, judging the estimated value of the heat absorption exceeding the normal working interval as abnormal, and releasing an alarm signal to the application service module;
the application service module is used for applying the estimated value of the heat absorption capacity to energy-saving control and big data analysis of the refrigerating system.
2. The system for estimating heat received by a refrigerant in a chiller according to claim 1 wherein the preprocessing in the data preprocessing module is performed by:
Detecting abnormal values of the operation parameters of the refrigeration system at each sampling moment by adopting a3 sigma criterion, and filling the abnormal values; performing interpolation estimation on the operation parameters of the refrigeration system and the missing values in the physical parameters of the refrigerant; further adopting a maximum and minimum value normalization method to perform standardization treatment; and taking the time stamp of the temperature data as a reference, and carrying out linear interpolation on data except the temperature data in the operation parameters of the refrigerating system and the physical parameters of the refrigerant so as to align the time stamp with the temperature data.
3. The system for estimating heat of refrigerant reception in a refrigerator according to claim 1, wherein the heat absorption capacity calculation module calculates the heat absorption capacity of the refrigerant of the compressor, the condenser, the evaporator, respectively, by a steady-state working fluid flow energy balance equation based on the preprocessed time-series data, and takes the heat absorption capacity of the compressor, the heat absorption capacity of the condenser, and the heat absorption capacity of the evaporator as initial estimated values of the heat absorption capacity.
4. The system for estimating heat of refrigerant in a refrigerator according to claim 3 wherein the neural network correction module corrects the initial estimate of heat absorption using a three-layer BP neural network, training of the BP neural network uses a gradient descent algorithm, and estimates the overall heat absorption estimate using the trained BP neural network based on compressor heat absorption, condenser heat release, evaporator heat absorption.
5. The system for estimating heat received by a refrigerant in a chiller according to claim 4 wherein the network structure of the BP neural network is as follows:
Input layer: 3 neurons for respectively receiving heat absorption capacity of the compressor, heat release capacity of the condenser and heat absorption capacity of the evaporator;
Hidden layer: 10 neurons, activation function is ReLU;
Output layer: and 1 neuron, outputting the corrected estimated heat absorption quantity value of the whole machine.
6. The system for estimating an amount of heat received by a refrigerant in a chiller according to claim 1 wherein the anomaly determination module determines a normal operating interval for the estimated amount of heat absorption based on a design specification of the chiller system and operating constraints.
7. The system for estimating an amount of heat received by a refrigerant in a refrigerator according to claim 1, wherein in the aspect of energy saving control in the application service module, the adjustment amounts of the compressor frequency and the expansion valve opening degree are calculated by a PID control algorithm based on the estimated amount of heat absorption in combination with a control target of the refrigeration system.
8. The system for estimating an amount of heat received by a refrigerant in a refrigerator according to claim 7, wherein when the estimated amount of heat absorption is smaller than the target amount of heat absorption, the compressor frequency is appropriately increased or the opening degree of the expansion valve is increased; when the estimated value of the heat absorption amount is larger than the target heat absorption amount, the compressor frequency should be appropriately decreased or the expansion valve opening degree should be decreased.
9. The system for estimating heat received by refrigerant in a chiller according to claim 1 wherein in terms of big data analysis, the application service module uploads the estimated value of heat absorption per sampling period, the compressor frequency, the expansion valve opening data and the alarm signal to the cloud platform, and the cloud platform stores and manages the data using the time series database after receiving the data.
10. A method for estimating the heat received by a refrigerant in a refrigerator, characterized by implementing a system for estimating the heat received by a refrigerant in a refrigerator as claimed in any one of claims 1 to 9, comprising the steps of:
Firstly, operation parameters of a refrigerating system and physical property parameters of a refrigerant are collected and preprocessed in real time to obtain preprocessed time series data;
further, combining the preprocessed time series data, calculating the heat absorption and release quantity of the refrigerant in the compressor, the condenser and the evaporator to obtain an initial estimated value of the heat absorption quantity; correcting the initial estimated value of the heat absorption capacity to obtain an estimated value of the heat absorption capacity;
further, defining a normal working interval of the estimated value of the heat absorption quantity, and judging whether the estimated value of the heat absorption quantity is in the normal working interval or not;
And finally, applying the estimated value of the heat absorption capacity to energy-saving control and big data analysis of the refrigerating system, and optimizing a control strategy of the refrigerating system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410666283.7A CN118296305A (en) | 2024-05-28 | 2024-05-28 | Method and system for estimating heat received by refrigerant in refrigerator |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410666283.7A CN118296305A (en) | 2024-05-28 | 2024-05-28 | Method and system for estimating heat received by refrigerant in refrigerator |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118296305A true CN118296305A (en) | 2024-07-05 |
Family
ID=91674488
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410666283.7A Pending CN118296305A (en) | 2024-05-28 | 2024-05-28 | Method and system for estimating heat received by refrigerant in refrigerator |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118296305A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105610959A (en) * | 2016-01-14 | 2016-05-25 | 中电长城网际***应用有限公司 | Information publishing-subscribing platform, method and system |
| CN111077870A (en) * | 2020-01-06 | 2020-04-28 | 浙江中烟工业有限责任公司 | Intelligent OPC data real-time acquisition and monitoring system and method based on stream calculation |
| CN111473471A (en) * | 2020-03-24 | 2020-07-31 | 珠海格力电器股份有限公司 | Multi-linkage energy consumption metering method and system |
| CN114396733A (en) * | 2021-12-31 | 2022-04-26 | 北京京仪自动化装备技术股份有限公司 | Control method of temperature control system and temperature control system |
| CN115235050A (en) * | 2022-09-21 | 2022-10-25 | 江苏橙智云信息技术有限公司 | Simulation method and device for energy-saving strategy of central air-conditioning water chilling unit |
| CN116045447A (en) * | 2023-01-03 | 2023-05-02 | 青岛海尔空调器有限总公司 | Air conditioner refrigeration control method and device and air conditioner |
| KR20230126044A (en) * | 2022-02-22 | 2023-08-29 | 엘지전자 주식회사 | Air conditioner |
-
2024
- 2024-05-28 CN CN202410666283.7A patent/CN118296305A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105610959A (en) * | 2016-01-14 | 2016-05-25 | 中电长城网际***应用有限公司 | Information publishing-subscribing platform, method and system |
| CN111077870A (en) * | 2020-01-06 | 2020-04-28 | 浙江中烟工业有限责任公司 | Intelligent OPC data real-time acquisition and monitoring system and method based on stream calculation |
| CN111473471A (en) * | 2020-03-24 | 2020-07-31 | 珠海格力电器股份有限公司 | Multi-linkage energy consumption metering method and system |
| CN114396733A (en) * | 2021-12-31 | 2022-04-26 | 北京京仪自动化装备技术股份有限公司 | Control method of temperature control system and temperature control system |
| KR20230126044A (en) * | 2022-02-22 | 2023-08-29 | 엘지전자 주식회사 | Air conditioner |
| CN115235050A (en) * | 2022-09-21 | 2022-10-25 | 江苏橙智云信息技术有限公司 | Simulation method and device for energy-saving strategy of central air-conditioning water chilling unit |
| CN116045447A (en) * | 2023-01-03 | 2023-05-02 | 青岛海尔空调器有限总公司 | Air conditioner refrigeration control method and device and air conditioner |
Non-Patent Citations (1)
| Title |
|---|
| 梁勇超，等: "空调器实时制冷量计算模型的研究与应用", 制冷与空调, 31 March 2024 (2024-03-31), pages 50 - 54 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102486390B (en) | Method for steam pipe network of iron and steel enterprise to correct metering data | |
| US7159408B2 (en) | Charge loss detection and prognostics for multi-modular split systems | |
| CN108709287B (en) | Air conditioning system cold load prediction method and water chilling unit group control strategy | |
| US11561019B2 (en) | Performance diagnosis device and performance diagnosis method for air conditioner | |
| CN113158494A (en) | Heat exchanger virtual-real fusion fault diagnosis method and system | |
| CN109711049B (en) | Hybrid subway air conditioner water chilling unit energy efficiency estimation method | |
| CN110688740A (en) | Modelica combined simulation optimization-based cold water machine room model calibration method | |
| WO2022025162A1 (en) | Abnormality diagnosis device, abnormality diagnosis method, and program | |
| CN118296305A (en) | Method and system for estimating heat received by refrigerant in refrigerator | |
| CN106840733B (en) | Air conditioning unit performance online detection method and detection device thereof | |
| CN117091273A (en) | Control method and device for self-adaptive learning data model of central air conditioning system | |
| TWI470217B (en) | On - line performance evaluation method of cooling tower | |
| CN102645315A (en) | Automatic, fast and accurate detection method for air resistance characteristics of large heat exchanger | |
| CN103245481B (en) | Frequency conversion technology based detection method for air resistance characteristic of large-scale variable-load heat exchanger | |
| TW201833518A (en) | Method of real-time prognosis of flooding phenomenon in packed columns | |
| KR100466433B1 (en) | Method for analysis of efficiency in turbo refrigerator and apparatus thereof | |
| JP2022044935A (en) | Energy plant performance estimation device | |
| CN114322321B (en) | Gray prediction extension constant temperature control method for gas water heater | |
| CN110779249A (en) | On-line diagnosis method for sensor concurrent faults in solar heat pump system | |
| WO2022025145A1 (en) | Estimation device, estimation method, and program | |
| CN117804112B (en) | Cold and heat source system Ai energy efficiency management system | |
| CN116447709A (en) | Heating ventilation air conditioner load prediction analysis system based on data analysis | |
| JP6948756B1 (en) | Performance diagnosis method for absorption chiller-heater | |
| WO2023279778A1 (en) | Control method and control apparatus for air conditioner, and air conditioner and readable storage medium | |
| CN114017908B (en) | Air conditioner tail end air outlet cold quantity control method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |