CN115343087A - Laboratory ventilation equipment's failure prediction system based on data analysis - Google Patents
Laboratory ventilation equipment's failure prediction system based on data analysis Download PDFInfo
- Publication number
- CN115343087A CN115343087A CN202210963756.0A CN202210963756A CN115343087A CN 115343087 A CN115343087 A CN 115343087A CN 202210963756 A CN202210963756 A CN 202210963756A CN 115343087 A CN115343087 A CN 115343087A
- Authority
- CN
- China
- Prior art keywords
- gas
- ventilation equipment
- emission
- fault
- corrosion
- 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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M99/00—Subject matter not provided for in other groups of this subclass
- G01M99/008—Subject matter not provided for in other groups of this subclass by doing functionality tests
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Testing Resistance To Weather, Investigating Materials By Mechanical Methods (AREA)
Abstract
The invention discloses a data analysis-based fault prediction system for laboratory ventilation equipment, which comprises an emission operation acquisition module, an environmental parameter acquisition module, an emission environmental corrosion analysis module, an emission amount analysis module, a data fusion analysis module, an emission amount analysis module and a data service platform. The method can analyze the corrosion predicted accelerated damage coefficient of each gas type discharged to the ventilation equipment, predict the predicted time length corresponding to the ventilation equipment reaching the preset air exhaust pipeline air leakage aperture threshold value from the current state by adopting a data prediction fusion formula, and analyze the error offset corresponding to the predicted time length by combining a data service platform so as to dynamically compensate the predicted time length, thereby improving the accuracy of the predicted time length of the continuous gas discharge of the ventilation equipment and improving the safety of the laboratory environment.
Description
Technical Field
The invention belongs to the technical field of laboratory ventilation equipment, and relates to a fault prediction system of laboratory ventilation equipment based on data analysis.
Background
In the routine analysis and experiment process of a laboratory, various pollution sources such as bad smelly, corrosive, toxic and harmful or inflammable and explosive gases and the like are often generated, harmful gases and solid particles are generated and are exhausted outdoors in time, indoor environment air pollution is caused, the health and safety of experimenters are influenced, and the precision and the service life of analytical instruments and equipment are influenced. The common ventilation equipment in the laboratory is a fume hood which is common exhaust equipment in a physical and chemical laboratory, and aims to control, dilute and exhaust toxic and harmful substances such as smoke, aerial fog and particles generated in the experiment, acidic, alkaline and corrosive gases and the like, and protect laboratory researchers from being damaged by the toxic and harmful chemical substances generated in the experiment.
Ventilation equipment in the current laboratory discharges along with acidity, basicity and corrosive gas, can lead to air pipe inner wall to receive the corruption, and then develop into air pipe corruption bore crescent, influence the air that ventilation equipment produced the laboratory and effectively discharge, and then lead to the unable in time elimination of harmful substance that produces in the experimentation, crisis experimenter's health and laboratory appliance's security, and simultaneously, when carrying out the failure prediction to ventilation equipment, can't get rid of the interference that causes each trouble room of ventilation equipment.
Disclosure of Invention
The invention aims to provide a fault prediction system of laboratory ventilation equipment based on data analysis, which solves the problems in the prior art.
The purpose of the invention can be realized by the following technical scheme:
a data analysis-based failure prediction system for laboratory ventilation equipment comprises an emission operation acquisition module, an environmental parameter acquisition module, an emission environment corrosion analysis module, an emission analysis module, a data fusion analysis module, an emission analysis module and a data service platform;
the emission operation acquisition module acquires operation parameters of each ventilation device in the laboratory and sends the acquired operation parameters of each ventilation device to the emission analysis module;
the environment parameter acquisition module comprises a plurality of environment parameter acquisition units, and the environment parameter acquisition units acquire the concentration of each gas type in the environment in the area in real time;
the exhaust environment corrosion analysis module extracts gas concentration and exhaust time corresponding to each gas type which is cumulatively exhausted by each ventilation device and humidity and temperature values in the ventilation device corresponding to each gas type which is exhausted, analyzes corrosion estimated accelerated damage coefficients of each gas type of the ventilation device in the exhaust environment to the ventilation device, and sends the analyzed corrosion estimated accelerated damage coefficients of each gas type to the ventilation device to the data fusion analysis module;
the emission analysis module acquires the operation parameters of the ventilation equipment in each region, which are acquired by the emission operation acquisition module, extracts the working power of the ventilation equipment in the operation parameters, and screens out the theoretical emission gas quantity LQ corresponding to the ventilation equipment within unit time T of power according to the working power of the ventilation equipment;
the data fusion analysis module extracts corrosion predicted accelerated damage coefficients of various gas types to the ventilation equipment, carries out data prediction fusion processing on the corrosion predicted accelerated damage coefficients of various gas types to the ventilation equipment and the time length of various gas types discharged by a laboratory in the past, predicts the predicted time length corresponding to the ventilation equipment reaching the set ventilation pipeline air leakage aperture threshold value from the current state, and feeds back the analyzed predicted time length to the data service platform;
the data service platform extracts the predicted duration which is sent by the data fusion analysis module and corresponds to the ventilation equipment reaching the set exhaust pipeline air leakage aperture threshold value from the current state, extracts the discharge duration and the discharged gas concentration of each gas type in the sampling time period in real time, calculates the error offset of the predicted duration by combining the average value SQ of the actual discharged gas quantity of the ventilation equipment in the sampling time period, and dynamically adjusts the predicted duration by adopting the error offset.
Further, the analysis of the predicted corrosion damage coefficient of each gas type to the ventilation equipment in a fixed monitoring time period by the emission environment corrosion analysis module comprises the following steps:
a1, extracting gas concentration of acidic, alkaline and corrosive gases of ventilation equipment when the gases reach discharge conditions, and temperature, humidity and discharge time in a discharge process to establish a gas discharge matrix L;
step A2, extracting the gas concentration under each gas type recorded in the gas emission matrix and the time required for finishing gas emission, and analyzing the decay rate of the gas emission concentration;
and A3, analyzing the corrosion prediction accelerated damage coefficient of each gas type discharged by the ventilation equipment to the ventilation equipment by adopting a corrosion damage prediction model.
Further, the formula of the corrosion damage estimation model is as follows:
pi is the corrosion estimated accelerated damage coefficient of the ventilation equipment under the interference of the ith exhaust gas, ci is the unit gas concentration of the ith gas species, unit mg/m3, beta i is the corrosion proportion coefficient corresponding to the unit gas concentration of the ith gas species, unit m3/mg, wp is the set standard humidity, wb is the set standard temperature, xi1 is the gas concentration corresponding to the ith gas species under the condition of reaching the discharge of the ventilation equipment, xi4 is the time required by the ith gas species discharged through the ventilation equipment, and Vi is the discharge concentration decay rate of the ith gas species.
Further, the emission analysis module extracts the wind speed V in the operation parameters, analyzes the actual amount of gas discharged by the ventilation equipment in unit time by adopting a wind volume calculation formula, and analyzes the theoretical amount of gas discharged and the actual amount of gas discharged SQ to obtain a gas emission attenuation coefficient.
Further, the data prediction fusion formula is adopted in the process of the data prediction fusion processing of the data fusion analysis module:
RS is the average value of theoretical exhaust gas quantity of the ventilation equipment under different working frequencies, WS is the average value of actual exhaust gas quantity of the ventilation equipment under different working frequencies at the current moment, XS is the average value of actual exhaust gas quantity under different working frequencies when the ventilation equipment reaches a set exhaust pipeline air leakage aperture threshold value, so that the large deviation degree of predicted duration caused by different working frequencies of the ventilation equipment is eliminated, the prediction of the gas sustainable exhaust duration is improved, E is the predicted duration of the sustainable exhaust gas, E is a natural number, and ui is the exhaust of the ith gas type in a previous laboratoryThe total length of time.
Further, an error offset calculation formula of the predicted time length is as follows:
e is a natural number, yi is the concentration of the gas required to be discharged by the ith gas species in the sampling time period, bi is the discharge duration of the ith gas species in the sampling time period, vi is the predicted average discharge rate of the ith gas species discharged by the ventilation equipment in the past when the duration E of the gas continuously discharged is obtained,
error offset coefficient for the predicted duration.
Furthermore, the fault prediction system of the laboratory ventilation equipment further comprises a fault emission training module, wherein the fault emission training module extracts actual emission gas quantities of the ventilation equipment at different working frequencies under different fault types and theoretical emission gas quantities under different working frequencies, performs dispersion training on the actual emission gas quantities under the fault types under the same working frequency, obtains standardization of the actual emission gas quantities after dispersion training, and analyzes emission output interference measures corresponding to the fault types.
Further, the actual exhaust gas amounts of different fault types under the same working frequency are subjected to decentralized training, and the specific training method comprises the following steps:
step 1, extracting actual exhaust gas quantities of different fault types under the same working frequency, wherein the times of extracting the actual exhaust gas quantities of the fault types are the same, namely the times of extracting samples of the same fault type under the same working frequency are f;
step 2, establishing a sample training feature table;
step 3, standardizing the actual discharged gas amount corresponding to each fault type to obtain a standardized training sample BQ j ，
k is the total number of samples, k = C N f, C is the operation of the ventilation deviceThe number of frequencies, i.e. frequency 1, frequency 2, frequency 3, the number of times that C takes the value 3, N is the fault category, SQ j Training the actual amount of the discharged gas corresponding to the jth sample in the feature table for the sample;
step 4, extracting standard training samples corresponding to different fault types under the same working frequency;
and 5, analyzing the emission output interference measure corresponding to each fault type by adopting an emission output interference measure model, and accurately eliminating the dynamic interference degree of the fault type on the air exhaust condition of the ventilation equipment under the interference of the working frequency.
Further, the formula of the emission output interference metric model in the step 5 is as follows:
a(BQ c ) Output of a disturbance variable, eta, for the exhaust air corresponding to the c-th fault category c Is the exhaust weight proportion coefficient corresponding to the c-th fault category,
z is equal to 1,2,3, c takes a value of 1,2, N, which is the actual amount of gas discharged corresponding to the c fault type of the ventilation equipment at the z working frequency,
actual amount of exhaust gas corresponding to the c-th fault category at the z-th operating frequency for the ventilation equipment, an
The invention has the beneficial effects that:
the invention analyzes the concentration and the discharge time of each gas type cumulatively discharged by the ventilation equipment and the temperature and the humidity in the ventilation equipment in the discharge process to obtain the corrosion estimated accelerated damage coefficient of each gas type discharged to the ventilation equipment, and can directly obtain the corrosion damage degree of each gas to the ventilation equipment in the discharge process.
The invention carries out comprehensive estimation and analysis on the corrosion estimated accelerated damage coefficient of the ventilation equipment and the time and the exhaust gas quantity of the ventilation equipment for each gas type discharged in the past by adopting a data prediction fusion formula in the process of discharging each gas type, estimates the predicted time length corresponding to the condition that the current state of the ventilation equipment reaches the preset air exhaust pipeline air leakage aperture threshold value, and analyzes the error offset corresponding to the predicted time length by combining a data service platform so as to carry out dynamic compensation on the predicted time length, thereby improving the accuracy of the predicted time length of the continuous gas discharge of the ventilation equipment, being capable of timely maintaining the ventilation pipeline once the predicted time length is close to the predicted time length, reducing the problem that harmful gas cannot be effectively discharged due to the corrosion damage of the pipeline, and improving the safety of the laboratory environment.
The method analyzes the emission output interference measure corresponding to each fault type through the fault emission training module, reduces the influence of different fault types on the dispersion of the actual emission gas volume under the same working frequency, and can accurately analyze the emission output interference measure corresponding to each fault type so as to avoid the dynamic interference of each fault type on the air exhaust of the ventilation equipment caused by different working frequencies and reduce the accuracy of the prediction of each fault type.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
A laboratory ventilation equipment fault prediction system based on data analysis comprises an emission operation acquisition module, an environmental parameter acquisition module, an emission environment corrosion analysis module, an emission amount analysis module, a data fusion analysis module, an emission amount analysis module, a data service platform and a fault emission training module.
The emission operation acquisition module acquires operation parameters of each ventilation device in the laboratory and sends the acquired operation parameters of each ventilation device to the emission analysis module.
The operating parameters of the ventilation equipment include wind pressure, wind speed, ventilation equipment frequency, and temperature and humidity inside the ventilation equipment.
The environment parameter acquisition module comprises a plurality of environment parameter acquisition units, and the environment parameter acquisition units acquire the concentration of each gas type in the environment in the area in real time.
The environment parameter acquisition units are distributed in the areas one by one, ventilation equipment and the environment parameter acquisition units are arranged in each area, the mutual mapping relation between the same ventilation equipment and the environment parameter acquisition units is established, and the selected ventilation equipment is a fume hood.
The collected gas species include carbon dioxide, chlorine, ammonia, nitrogen oxides, hydrogen fluoride, sulfur dioxide, hydrogen sulfide, carbon monoxide, methane, acetylene, ethylene, propane, butane, propylene, propyne, 1, 3-butadiene, butyne, hydrogen sulfide, and the like.
The exhaust environment corrosion analysis module extracts gas concentration and exhaust time corresponding to each gas type which is cumulatively exhausted by each ventilation device and humidity and temperature values in the ventilation device corresponding to each gas type which is exhausted, analyzes corrosion prediction accelerated damage coefficients of each gas type of the ventilation device in the exhaust environment to the ventilation device, and sends the analyzed corrosion prediction accelerated damage coefficients of each gas type to the ventilation device to the data fusion analysis module.
The damage coefficient is estimated according to the corrosion of various gas types which are accumulatively discharged by the ventilation equipment to the ventilation equipment, so that when acid, alkaline and corrosive gases generated in a laboratory are judged to pass through the ventilation equipment and a ventilation equipment pipeline, the ventilation equipment is corroded and damaged, the ventilation equipment leaks air when in use along with the corrosion of the ventilation equipment, and harmful gases generated in the laboratory cannot be stably and effectively discharged.
Specifically, the analysis of the predicted corrosion damage coefficient of each gas type to the ventilation equipment in the fixed monitoring time period by the emission environment corrosion analysis module comprises the following steps:
step A1, extracting the gas concentration of the ventilation equipment when the acidic, alkaline and corrosive gases reach the discharge condition, and the temperature, humidity and discharge time in the discharge process, and establishing a gas discharge matrix
i =1,2,.. M, m is the total number of times of accumulating the gas types discharged through the ventilation equipment according to the time sequence, xi1 is the gas concentration corresponding to the ith gas type under the condition of reaching the discharge of the ventilation equipment, xi2 is the humidity in the ventilation equipment during the discharge of the ith gas type discharged through the ventilation equipment, xi3 is the temperature in the ventilation equipment during the discharge of the ith gas type discharged through the ventilation equipment, and xi4 is the time required by the ith gas type discharged through the ventilation equipment.
Step A2, extracting the gas concentration xi1 under each gas type recorded in the gas emission matrix and the time xi4 required by the gas emission, and analyzing the decay rate of the gas emission concentration
Step A3, analyzing corrosion prediction accelerated damage coefficients of various gas types discharged by the ventilation equipment to the ventilation equipment by adopting a corrosion damage prediction model;
the formula of the corrosion damage estimation model is as follows:
pi is the corrosion estimated accelerated damage coefficient of the ventilation equipment under the interference of the ith exhaust gas, ci is the unit gas concentration of the ith gas species, unit mg/m3, beta i is the corrosion proportion coefficient corresponding to the unit gas concentration of the ith gas species, unit m3/mg, wp is the set standard humidity, and wb is the set standard temperature.
Through analyzing the environmental parameters in the ventilation equipment and the concentration of each gas discharged when each gas type that produces in the laboratory is discharged through ventilation equipment, the corrosivity of discharging and the corruption aggravation damage degree of acid and alkali gas to ventilation equipment can be predicted, realize the concrete numerical degree quantization show of single gas type to ventilation pipe damage, and along with corrosivity, the corrosive action of acid and alkali gas under special environment can make pipeline and the ventilation equipment body (or the check valve block in the ventilation equipment) among the ventilation equipment corrode and drop, lead to ventilation equipment to leak gas, influence the discharge efficiency of the harmful gas that produces in the laboratory.
The emission analysis module acquires the operation parameters of the ventilation equipment in each region acquired by the emission operation acquisition module, extracts the working power of the ventilation equipment in the operation parameters, screens out the theoretical discharged gas quantity LQ corresponding to the power of the ventilation equipment in unit time T according to the working power of the ventilation equipment, extracts the wind speed V in the operation parameters, analyzes the actual discharged gas quantity of the ventilation equipment in unit time by adopting a wind quantity calculation formula, analyzes the theoretical discharged gas quantity and the actual discharged gas quantity SQ, and acquires a gas emission attenuation coefficient
By performing differential analysis on the theoretical discharge gas quantity and the actual discharge gas quantity, the discharge attenuation coefficient of the actual discharge gas quantity relative to the theoretical discharge gas quantity can be judged.
The actual gas discharge quantity SQ is calculated according to the formula SQ = V x D x T, D is the area of the opening of the fume hood operation window, V is the wind speed at the position of the operation window, and T is unit time.
The data fusion analysis module extracts corrosion predicted accelerated damage coefficients of all gas types to the ventilation equipment, carries out data prediction fusion processing on the corrosion predicted accelerated damage coefficients of all gas types to the ventilation equipment and the time length of all gas types discharged in the past in a laboratory, predicts the predicted time length corresponding to the current state of the ventilation equipment reaching the set air exhaust pipeline air leakage aperture threshold value (the set air exhaust pipeline air leakage aperture threshold value is the accumulated area of the pipeline air leakage aperture corresponding to the preset discharge lower limit numerical value of the discharge amount of the gas discharged by the ventilation equipment), and feeds back the analyzed predicted time length to the data service platform.
The data prediction fusion formula is adopted in the process of performing data prediction fusion processing by the data fusion analysis module:
RS is the average value of theoretical exhaust gas quantity of the ventilation equipment under different working frequencies, WS is the average value of actual exhaust gas quantity of the ventilation equipment under different working frequencies at the current moment, XS is the average value of actual exhaust gas quantity under different working frequencies when the ventilation equipment reaches a set exhaust pipeline air leakage aperture threshold value, so that the large deviation degree of the predicted duration caused by different working frequencies of the ventilation equipment is eliminated, the prediction of the gas sustainable exhaust duration is improved, E is the predicted duration of the sustainable exhaust gas, E is a natural number, and ui is the total discharge duration of the ith gas type in a conventional laboratory.
The method has the advantages that when the pipeline of the ventilation equipment reaches the set air exhaust pipeline air leakage aperture threshold value from the current state, the predicted time for the gas to be continuously exhausted can be predicted through the data fusion analysis module, the ventilation equipment does not have other faults in the process of predicting the time for the gas to be continuously exhausted of the ventilation equipment, and otherwise, the prediction accuracy can be influenced.
The data service platform extracts the predicted time length corresponding to the ventilation equipment reaching the set exhaust pipeline air leakage aperture threshold value from the current state, extracts the discharge time length of each gas type and the discharged gas concentration in a sampling time period in real time, calculates the error offset of the predicted time length by combining the average value SQ of the actual discharged gas quantity of the ventilation equipment in the sampling time period, dynamically adjusts the predicted time length by adopting the error offset, and adjusts the adjusted predicted time length to be the preset exhaust pipeline air leakage aperture threshold value
The accuracy of the prediction time length of the failure of the ventilation equipment is ensured, the adjustment is convenient according to the actual gas emission of the ventilation equipment in the sampling time period, the air leakage of the ventilation equipment pipeline can be visually displayed, and the manual work can be remindedAnd the pipeline is overhauled or replaced in time.
An error offset calculation formula of the predicted time length is as follows:
e is a natural number, yi is the concentration of the gas required to be discharged by the ith gas species in the sampling time period, bi is the discharge duration of the ith gas species in the sampling time period, vi is the predicted average discharge rate of the ith gas species discharged by the ventilation equipment in the past when the duration E of the gas continuously discharged is obtained,
error offset coefficient for the predicted duration.
The method comprises the steps of calculating the error offset of the predicted duration E by adopting an error offset calculation formula to obtain an error offset coefficient, dynamically adjusting the sustainable discharge duration of the ventilation equipment according to the error offset coefficient to continuously adjust the sustainable discharge duration, further optimizing the predicted duration, ensuring that the error between the dynamically adjusted predicted duration and the actual sustainable discharge duration is smaller, improving the accuracy of the predicted duration, timely maintaining the ventilation pipeline once the predicted duration is close to the predicted duration, reducing the problem that harmful gas cannot be effectively discharged due to corrosion damage of the pipeline, and improving the safety of the laboratory environment.
The fault emission training module extracts actual emission gas quantities of the ventilation equipment at different working frequencies under different fault types and theoretical emission gas quantities under different working frequencies, the actual emission gas quantities under the fault types under the same working frequency are subjected to dispersion training, standardization of the actual emission gas quantities after the dispersion training is obtained, emission output interference measures corresponding to the fault types are analyzed, sampling dispersion training can obtain influences of the different fault types under the same working frequency on the actual emission gas quantity dispersion, the interference of the working frequency is eliminated by adopting standardization processing, the emission output interference measures corresponding to the fault types can be accurately analyzed, dynamic interference of the fault types on ventilation equipment exhaust caused by the different working frequencies is avoided, and accuracy of prediction of the fault types is reduced.
The method comprises the following steps of performing dispersion training on actual exhaust gas quantities of different fault types under the same working frequency, wherein the specific training method comprises the following steps:
step 1, extracting actual exhaust gas quantities of different fault types at the same operating frequency, wherein the extracted actual exhaust gas quantities of the fault types are the same in frequency, namely the frequency of samples extracted for the same fault type at the same operating frequency is f.
The fault types comprise the conditions of air leakage of an exhaust pipeline, damage of a valve plate of a non-return air valve, fault of a fan, fault of a motor, abnormity of a frequency converter, unstable power supply voltage and the like.
For the fault types of air leakage of the air exhaust pipeline, in the process of dispersion training, state training is carried out according to different air leakage grades, the air leakage of the powder exhaust pipeline is divided into a first air leakage grade, a second air leakage grade and a third air leakage grade according to the set air leakage caliber of the air exhaust pipeline as a standard, the size of the air leakage port of the air exhaust pipeline is smaller than the set air leakage caliber of the air exhaust pipeline, the size of the air leakage port of the air exhaust pipeline is smaller than 2 times of the set air leakage caliber of the air exhaust pipeline and larger than the set air leakage port of the air exhaust pipeline, and the size of the air leakage port of the air exhaust pipeline is larger than 2 times of the set air leakage port caliber of the air exhaust pipeline.
Step 2, establishing a sample training feature table, namely recording a plurality of groups of sample training feature sub-tables in the sample training feature table, wherein each group of sample training feature sub-tables records the actual emission gas quantity corresponding to each fault type under the same working frequency and the theoretical emission gas quantity of the ventilation equipment under the working frequency, and sequencing the actual emission gas quantity and the theoretical emission gas quantity in sequence according to the working frequency from large to small;
step 3, standardizing the actual discharged gas amount corresponding to each fault type to obtain a standardized training sample BQ j ，
k is the total number of samples, k = C N f, C is the number of operating frequencies of the ventilation device, i.e. frequency 1, frequency 2, frequency 3, C takes the value 3N is the number of times of the fault type, SQ j Training the actual amount of the discharged gas corresponding to the jth sample in the feature table for the sample;
step 4, extracting standard training samples corresponding to different fault types under the same working frequency;
and 5, analyzing the emission output interference measure corresponding to each fault type by adopting an emission output interference measure model, and accurately eliminating the dynamic interference degree of the fault type on the air exhaust condition of the ventilation equipment under the interference of the working frequency.
Wherein the formula of the emission output disturbance metric model is:
a(BQ c ) Output of a disturbance variable, eta, for the exhaust air corresponding to the c-th fault category c Is the exhaust weight proportion coefficient corresponding to the c-th fault category,
z is equal to 1,2,3, c takes the value 1,2, N, which is the actual amount of exhaust gas corresponding to the c-th fault category of the ventilation equipment at the z-th operating frequency,
actual amount of exhaust gas corresponding to the c-th fault category at the z-th operating frequency for the ventilation equipment, an
Through analyzing the emission output interference measurement that each trouble kind corresponds, the influence that different trouble kinds brought the dispersion of actual exhaust gas volume under the same operating frequency can be acquireed in the scattered training of sampling, adopt standardized processing to get rid of operating frequency's interference, and then the emission output interference measurement that each trouble kind corresponds can be accurately analyzed out, dynamic interference to ventilation equipment is aired exhaust because of each trouble kind that operating frequency difference leads to is avoided, reduce the accuracy of each trouble kind prediction, simultaneously can carry out differentiation classification to each trouble kind.
The above formulas are all calculated by taking the numerical value of the dimension, the formula is a formula of the latest real situation obtained by collecting a large amount of data and performing software simulation, the preset parameters in the formula are set by the technical personnel in the field according to the actual situation, the scale coefficient and the weight coefficient are specific numerical values obtained by quantizing each parameter, and the subsequent comparison is convenient, and the scale coefficient and the weight coefficient can be calculated as long as the proportional relation between the parameter and the quantized numerical value is not influenced.
The foregoing is merely exemplary and illustrative of the principles of the present invention and various modifications, additions and substitutions of the specific embodiments described herein may be made by those skilled in the art without departing from the principles of the present invention or exceeding the scope of the claims set forth herein.
Claims (9)
1. A laboratory ventilation equipment failure prediction system based on data analysis, characterized by: the system comprises an emission operation acquisition module, an environmental parameter acquisition module, an emission environmental corrosion analysis module, an emission analysis module, a data fusion analysis module, an emission analysis module and a data service platform;
the emission operation acquisition module acquires operation parameters of each ventilation device in the laboratory and sends the acquired operation parameters of each ventilation device to the emission analysis module;
the environment parameter acquisition module comprises a plurality of environment parameter acquisition units, and the environment parameter acquisition units acquire the concentration of each gas type in the environment in the area in real time;
the exhaust environment corrosion analysis module extracts gas concentration and exhaust time corresponding to each gas type which is cumulatively exhausted by each ventilation device and humidity and temperature values in the ventilation device corresponding to each gas type which is exhausted, analyzes corrosion estimated accelerated damage coefficients of each gas type of the ventilation device to the ventilation device in the exhaust environment, and sends the analyzed corrosion estimated accelerated damage coefficients of each gas type to the ventilation device to the data fusion analysis module;
the emission analysis module acquires the operation parameters of the ventilation equipment in each region, which are acquired by the emission operation acquisition module, extracts the working power of the ventilation equipment in the operation parameters, and screens out the theoretical emission gas quantity LQ corresponding to the ventilation equipment within unit time T of power according to the working power of the ventilation equipment;
the data fusion analysis module extracts corrosion predicted accelerated damage coefficients of all gas types to the ventilation equipment, carries out data prediction fusion processing on the corrosion predicted accelerated damage coefficients of all gas types to the ventilation equipment and the duration of all gas types discharged in the past in a laboratory, predicts the predicted duration corresponding to the current state of the ventilation equipment reaching the set exhaust pipeline air leakage caliber threshold value, and feeds back the analyzed predicted duration to the data service platform;
the data service platform extracts the predicted time length corresponding to the ventilation equipment reaching the set air leakage caliber threshold value of the exhaust pipeline from the current state, extracts the discharge time length and the discharged gas concentration of each gas type in the sampling time period in real time, calculates the error offset of the predicted time length by combining the average value SQ of the actual discharged gas quantity of the ventilation equipment in the sampling time period, and dynamically adjusts the predicted time length by adopting the error offset.
2. The system of claim 1, wherein the laboratory ventilation equipment failure prediction system based on data analysis comprises: the analysis of the predicted corrosion damage coefficient of each gas type to the ventilation equipment in a fixed monitoring time period by the emission environment corrosion analysis module comprises the following steps:
a1, extracting gas concentration of acidic, alkaline and corrosive gases of ventilation equipment when the gases reach discharge conditions, and temperature, humidity and discharge time in a discharge process to establish a gas discharge matrix L;
step A2, extracting the gas concentration under each gas type recorded in the gas emission matrix and the time required by the gas emission completion, and analyzing the attenuation rate of the gas emission concentration;
and A3, analyzing the corrosion prediction accelerated damage coefficient of each gas type discharged by the ventilation equipment to the ventilation equipment by adopting a corrosion damage prediction model.
3. The system of claim 2, wherein the laboratory ventilation equipment failure prediction system based on data analysis comprises: the formula of the corrosion damage estimation model is as follows:
pi is the corrosion estimated accelerated damage coefficient of the ventilation equipment under the interference of the ith exhaust gas, ci is the unit gas concentration of the ith gas species, unit mg/m3, beta i is the corrosion proportion coefficient corresponding to the unit gas concentration of the ith gas species, unit m3/mg, wp is the set standard humidity, wb is the set standard temperature, xi1 is the gas concentration corresponding to the ith gas species under the condition of reaching the discharge of the ventilation equipment, xi4 is the time required by the ith gas species discharged through the ventilation equipment, and Vi is the discharge concentration decay rate of the ith gas species.
4. The system of claim 3, wherein the laboratory ventilation equipment fault prediction system based on data analysis comprises: the emission analysis module also extracts the wind speed V in the operation parameters, analyzes the actual amount of the discharged gas of the ventilation equipment in unit time by adopting a wind volume calculation formula, and analyzes the theoretical amount of the discharged gas and the actual amount SQ of the discharged gas to obtain a gas emission attenuation coefficient.
5. The system of claim 1, wherein the laboratory ventilation equipment failure prediction system based on data analysis comprises: the data prediction fusion formula is adopted in the process of performing data prediction fusion processing by the data fusion analysis module:
RS is the average value of theoretical exhaust gas quantity of the ventilation equipment under different working frequencies, WS is the current timeThe average value of the actual gas discharge quantity of the ventilation equipment under different working frequencies is obtained, XS is the average value of the actual gas discharge quantity under different working frequencies when the ventilation equipment reaches a set air leakage caliber threshold value of an air exhaust pipeline, so that the large deviation degree of the predicted duration caused by different working frequencies of the ventilation equipment is eliminated, the prediction of the duration of the gas continuous discharge is improved, E is the predicted duration of the gas continuous discharge, E is a natural number, and ui is the total discharge duration of the ith gas type in a conventional laboratory.
6. The system of claim 5, wherein the laboratory ventilation equipment failure prediction system based on data analysis comprises: the error offset of the predicted time length is calculated by the formula:
e is a natural number, yi is the concentration of the gas required to be discharged by the ith gas species in the sampling time period, bi is the discharge duration of the ith gas species in the sampling time period, vi is the predicted average discharge rate of the ith gas species discharged by the ventilation equipment in the past when the duration E of the gas continuously discharged is obtained,
error offset coefficient for the predicted duration.
7. The system of any one of claims 1-6 for predicting failure of a laboratory ventilation device based on data analysis, wherein: the fault prediction system of the laboratory ventilation equipment further comprises a fault emission training module, wherein the fault emission training module extracts actual emission gas quantities of the ventilation equipment with different working frequencies under different fault types and theoretical emission gas quantities under different working frequencies, performs dispersion training on the actual emission gas quantities under the fault types under the same working frequency, obtains standardization of the actual emission gas quantities after dispersion training, and analyzes emission output interference measures corresponding to the fault types.
8. The system of claim 7, wherein the laboratory ventilation equipment failure prediction system based on data analysis comprises: the method comprises the following steps of performing dispersion training on actual exhaust gas quantities of different fault types under the same working frequency, wherein the specific training method comprises the following steps:
step 1, extracting actual exhaust gas quantities of different fault types under the same working frequency, wherein the extracted actual exhaust gas quantities of the fault types are the same in times, namely the times of extracting samples of the same fault type under the same working frequency are f;
step 2, establishing a sample training feature table;
step 3, standardizing the actual discharged gas amount corresponding to each fault type to obtain a standardized training sample BQ j ，
k is the total number of samples, k = C N f, C is the number of operating frequencies of the ventilation device, i.e. frequency 1, frequency 2, frequency 3, C takes the value 3, N is the number of fault categories, SQ j Training the actual gas discharge amount corresponding to the jth sample in the characteristic table for the sample;
step 4, extracting standard training samples corresponding to different fault types under the same working frequency;
and 5, analyzing the emission output interference measure corresponding to each fault type by adopting an emission output interference measure model, and accurately eliminating the dynamic interference degree of the fault type on the air exhaust condition of the ventilation equipment under the interference of the working frequency.
9. The system of claim 8, wherein the laboratory ventilation equipment failure prediction system based on data analysis comprises: the formula of the emission output interference metric model in step 5:
a(BQ c ) Output of a disturbance variable, eta, for the exhaust air corresponding to the c-th fault category c For the c-th fault categoryThe corresponding air exhaust weight proportion coefficient is obtained,
z is equal to 1,2,3, c takes the value 1,2, N, which is the actual amount of exhaust gas corresponding to the c-th fault category of the ventilation equipment at the z-th operating frequency,
actual amount of exhaust gas corresponding to the c-th fault category at the z-th operating frequency for the ventilation equipment, an
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210963756.0A CN115343087A (en) | 2022-08-11 | 2022-08-11 | Laboratory ventilation equipment's failure prediction system based on data analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210963756.0A CN115343087A (en) | 2022-08-11 | 2022-08-11 | Laboratory ventilation equipment's failure prediction system based on data analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115343087A true CN115343087A (en) | 2022-11-15 |
Family
ID=83952968
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210963756.0A Pending CN115343087A (en) | 2022-08-11 | 2022-08-11 | Laboratory ventilation equipment's failure prediction system based on data analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115343087A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117110554A (en) * | 2023-10-24 | 2023-11-24 | 广东埃文低碳科技股份有限公司 | Thermal power enterprise greenhouse gas terminal emission monitoring system and method |
-
2022
- 2022-08-11 CN CN202210963756.0A patent/CN115343087A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117110554A (en) * | 2023-10-24 | 2023-11-24 | 广东埃文低碳科技股份有限公司 | Thermal power enterprise greenhouse gas terminal emission monitoring system and method |
| CN117110554B (en) * | 2023-10-24 | 2024-01-12 | 广东埃文低碳科技股份有限公司 | Thermal power enterprise greenhouse gas terminal emission monitoring system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111804146B (en) | Intelligent ammonia injection control method and intelligent ammonia injection control device | |
| CN109240244B (en) | Data-driven equipment running state health degree analysis method and system | |
| CN108956876B (en) | Measurement delay correction method for smoke on-line continuous monitoring system | |
| CN106248864B (en) | A kind of SCR denitration life-span prediction method based on magnanimity operation data | |
| CN109557869A (en) | A kind of fired power generating unit carbon emission on-line monitoring management system | |
| CN114354783A (en) | Health degree evaluation method of extra-high voltage oil chromatography monitoring device based on-operation data | |
| CN117148001A (en) | New energy automobile fills electric pile fault prediction system based on artificial intelligence | |
| CN116559684A (en) | Large-scale battery energy storage power station running state monitoring and evaluating system | |
| CN115343087A (en) | Laboratory ventilation equipment's failure prediction system based on data analysis | |
| CN109634944B (en) | Network loss data cleaning method based on multi-dimensional space-time analysis | |
| CN111753389B (en) | On-line soft measurement method and device for calorific value of raw coal fed into coal-fired boiler | |
| CN103942735A (en) | Method for evaluating relay protection states | |
| CN107631964A (en) | Haze filth fallout monitoring system and monitoring method | |
| CN116505663B (en) | Farm power consumption safety state monitoring and early warning system | |
| CN111612019A (en) | Method for identifying and analyzing fault abnormality of intelligent electric meter based on big data model | |
| CN116887569A (en) | Data center energy consumption prediction and energy saving adjustment method, system and storage medium | |
| CN115078924A (en) | System and method for judging transformer defects by combining dissolved gas in oil and iron core grounding current | |
| CN117849700B (en) | Modular electric energy metering system capable of controlling measurement | |
| CN117114519A (en) | Comprehensive energy operation system data calculation method | |
| CN113505497B (en) | Method and system for monitoring slurry quality of wet flue gas desulfurization absorption tower | |
| CN118152784A (en) | Modularized substation equipment data feature extraction method | |
| CN117035564B (en) | Construction quality supervision system suitable for engineering supervision | |
| CN114266454A (en) | Cigarette factory energy consumption abnormity monitoring and judging method and system thereof | |
| CN108775921A (en) | Industrial smoke on-line continuous monitoring device | |
| CN113390641A (en) | Intelligent early warning and online diagnosis method and system for equipment faults of wind and smoke system |
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 |