KR102095890B1 - Gas measuring system - Google Patents

Abstract

The present invention provides a gas measuring system which can calculate a partial pressure value minimizing a whole cost function. According to the present invention, the gas measuring system includes a calculation module inputting a measurement value of at least a gas sensor measured in mixed gas into the cost function, and calculating each of the partial pressure values of at least one gas, which minimizes the cost function. The calculation module includes: a learning unit calculating a correlation coefficient of the cost function minimizing the cost function by inputting the partial pressure value and the measurement value, which are measured under an allowed environment where the number of the gases and the partial pressure value are obtained, into the cost function, and repeating and learning a process of changing the number of the gases and the partial pressure value and calculating the correlation coefficient of the cost function minimizing the cost function; and a calculation unit calculating each partial pressure value of the gases to be measured by receiving the input of the measurement value of the plurality of gas sensors and the number and kind of the gases to be measured in the mixed gas in the cost function applied with the correlation coefficient learned by being calculated from the learning unit.

Description

Gas measurement system {GAS MEASURING SYSTEM}

The present invention relates to a gas measurement system.

In general, when entering a confined area where there is a risk of poisoning, fire explosion, etc. due to oxygen deficiency or harmful hazardous gas, the worker first enters after confirming whether it is safe by measuring the concentration of harmful gas inside.

In particular, in ships or offshore construction, it is stipulated to enter and exit the work area through measurement of complex gas (oxygen and combustible gas, etc.) for the closed area. However, the purchase cost and safety monitor are increasing factors for the installation of the gas detector individually for each closed area.

Looking at the conventional method for measuring the gas concentration in the closed area, the oxygen meter is fixedly installed at the bottom of the closed area and the oxygen concentration is checked at the entrance. That is, after installing the oxygen meter in the lower part of the sealed area, connect the cable to enter the sealed area, check the oxygen concentration in the upper part before going down to work.

However, in the case of such a conventional method, by measuring only a single oxygen gas concentration limited to a specific location in an enclosed area, an additional cost is incurred because an additional gas meter must be installed, if necessary. There is this.

In addition, it tends to be avoided by the operator because it is heavy when installing and dismantling the gas meter. That is, when installing the gas meter, the gas meter and the cable must be lowered and installed under the closed area, so that the worker avoids installing the gas meter and causes a decrease in work efficiency.

An object of the present invention is to provide a gas measurement system capable of calculating a partial pressure value that minimizes the overall cost function after learning an optimized relationship coefficient between a gas partial pressure (concentration) and a gas sensor measurement value.

On the other hand, the technical problems to be achieved in the present invention are not limited to the technical problems mentioned above, and other technical problems that are not mentioned are clearly apparent to those skilled in the art from the following description. Will be understandable.

A gas measurement system according to an embodiment of the present invention is a gas measurement system for measuring a partial pressure value of a specific gas using a measurement value measured by a plurality of gas sensors in a mixed gas, wherein at least one gas measured in the mixed gas A calculation module for inputting a measurement value of a sensor into a cost function and calculating each partial pressure value of at least one gas that minimizes the cost function; Including, the calculation module calculates the correlation coefficient of the cost function to minimize the cost function by inputting the measured value and the partial pressure value in a reference environment in which the number and partial pressure value of each gas are known, and each A learning unit learning this by repeatedly performing a process of calculating a correlation coefficient of a cost function that changes the number of gas and partial pressure values and minimizes the cost function; And the type, number of gases to be measured in the mixed gas and the measured values of a plurality of gas sensors are input to a cost function to which a correlation coefficient calculated and learned by the learning unit is applied to calculate each partial pressure value of the gas to be measured. Calculation unit; It includes.

In addition, the correlation coefficient includes a first correlation coefficient and a second correlation coefficient, and the first correlation coefficient is a correlation coefficient for a correlation between a partial pressure value of each of the at least one gas and a measurement value of the at least one gas sensor. In addition, the second correlation coefficient may be a correlation coefficient for a correlation between partial pressure values of a plurality of gases.

를 통해 상기 비용 함수를 최소화하는 비용 함수의 상관 계수를 산출하며, 여기서, α는 제1상관 계수, δ는 제2상관 계수, M은 가스 센서의 개수, N은 가스환경 실험조건에 따른 센서특성 학습용 데이터 개수, P는 측정하고자 하는 가스의 개수, S는 가스 센서의 측정값, C는 가스의 분압값, l은 1 내지 P 중 어느 하나의 자연수를 나타낼 수 있다.In addition, the learning unit

Calculate the correlation coefficient of the cost function to minimize the cost function, where α is the first correlation coefficient, δ is the second correlation coefficient, M is the number of gas sensors, and N is the sensor characteristics according to the gas environment experimental conditions. The number of learning data, P is the number of gases to be measured, S is the measured value of the gas sensor, C is the partial pressure value of the gas, and l can represent any natural number from 1 to P.

를 통해 상기 비용 함수를 최소화하는 적어도 하나의 가스의 분압값을 산출하며, 여기서, C_jk는 산출하는 목표값, α는 제1상관 계수, δ는 제2상관 계수, M은 가스 센서의 개수, P는 측정하고자 하는 가스의 개수, S는 가스 센서의 측정값을 나타낼 수 있다.In addition, the calculation unit

To calculate the partial pressure value of at least one gas minimizing the cost function, C _jk is a target value to be calculated, α is a first correlation coefficient, δ is a second correlation coefficient, M is the number of gas sensors, P may indicate the number of gases to be measured, and S may indicate the measured value of the gas sensor.

In addition, a sensor module including the plurality of gas sensors and a communication module transmitting measurement values measured by the sensor module to the calculation module may be further included.

In addition, a protection member surrounding the sensor module may be further included, and an inflow hole for introducing the gas may be formed in the protection member.

In addition, it is disposed inside the protective member, it may further include a renewable energy generating unit for supplying power to the sensor module.

According to an embodiment of the present invention, it is possible to calculate the partial pressure value that minimizes the overall cost function after learning the optimized relationship coefficient between the gas partial pressure (concentration) and the gas sensor measurement value.

On the other hand, the effects obtainable in the present invention are not limited to the above-mentioned effects, and other effects not mentioned will be clearly understood by those skilled in the art from the following description. Will be able to.

1 is a block diagram schematically showing a gas measurement system according to an embodiment of the present invention,
2 is an exemplary view schematically showing a sensor module in a gas measurement system according to an embodiment of the present invention.
3 is an exemplary view showing a state in which a sensor module is accommodated in a protective member in a gas measurement system according to another embodiment of the present invention.
4 is an exemplary view showing a state in which a renewable energy generating unit is attached to a protection member in a gas measurement system according to another embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in more detail with reference to the accompanying drawings. The embodiments of the present invention can be modified in various forms, and the scope of the present invention should not be interpreted as being limited to the following embodiments. This embodiment is provided to more fully describe the present invention to those skilled in the art. Therefore, the shape of the elements in the drawings has been exaggerated to emphasize a clearer explanation.

The configuration of the invention for clarifying the solution to the problem to be solved by the present invention will be described in detail with reference to the accompanying drawings based on preferred embodiments of the present invention, but the same is used in assigning reference numbers to components of the drawings. For the components, even if they are on different drawings, the same reference numbers are assigned, and it is revealed in advance that components of other drawings may be cited when necessary for the description of the drawings.

1 is a block diagram schematically showing a gas measurement system according to an embodiment of the present invention, and FIG. 2 is an exemplary view schematically showing a sensor module in a gas measurement system according to an embodiment of the present invention.

1 and 2, a gas measurement system according to an embodiment of the present invention may include a sensor module 100, a communication module 200, and a calculation module 300.

The sensor module 100 may be exposed to a mixed gas environment, and measure and analyze the type and partial pressure information of each gas present in the mixed gas.

The sensor module 100 may have a plurality of gas sensors arranged in an array form on a single or multiple substrates.

The sensor module 100 may collect and manage measurement values measured by each gas sensor.

Each gas sensor uses an interaction between a gas and a substance, and is divided into various types according to the detection method.

Each gas sensor may include a semiconductor gas sensor, a solid electrolyte gas sensor, an electrochemical gas sensor, a contact combustion gas sensor, and a non-dispersive infrared gas sensor according to the detection method.

Here, the semiconductor type gas sensor has the advantage of being able to be manufactured in a low-cost, low-power and small size, and has a long life, but has a disadvantage of low selectivity, and the electrochemical formula has excellent selectivity, but has a short life and short expensive manufacturing. , Catalytic combustion type and solid electrolyte type are only applicable to the detection of specific gas due to the small amount of gas that can be measured. The non-dispersive infrared type, which is an optical measurement method, increases the complexity of the measurement system due to large interference by other components.

Due to these characteristics, the gas sensor can be selected according to the gas to be measured and the application purpose.

The communication module 200 may be connected to the sensor module 100 or formed integrally to transmit measurement data measured by the sensor module 100 to the calculation module 300.

Here, the communication module 200 may include a transmission / reception module connected to the sensor module 100 and a transmission / reception module connected to the calculation module 300.

Here, the communication module 200 may include a short-range communication module or a long-range communication module.

The short-range communication module refers to a module for short-range communication.

As an example, the short-range communication module includes NFC (Near Field Communication) communication, Bluetooth communication, Radio Frequency Identification (RFID) communication, IrDA (Infrared Data Association) communication, UWB (Ultra-Wideband) communication, Zigbee communication, LoRa communication, RADAR communication , It may be a communication module that performs wireless communication according to any one of known short-range wireless communication methods such as low-power wireless communication and WiFi communication.

In addition, the telecommunication module refers to a module for telecommunication.

The module for telecommunications is a module for telecommunications that performs wireless communication according to any one of known telecommunications methods such as GSM, GPRS, EDGE, LTE-A, LTE, CDMA, WCDMA, EVDO, Wibro, Mobile WiMax, etc. You can.

The calculation module 300 may receive measurement data measured by the sensor module 100 transmitted through the communication module 200.

The calculation module 300 may include a learning unit 310 and a calculation unit 320.

Here, the learning unit 310 and the calculation unit 320 may be implemented as a single control algorithm.

The learning unit 310 receives the measured value and the partial pressure value of the gas sensor measured in a given environment in which the number and partial pressure value of each gas are known, and receives the specific data from the cost function defined by Equation 1 below. , It can be learned by repeating the process of calculating the correlation coefficient of the cost function to minimize the cost function, changing the number and partial pressure value of each gas, and calculating the correlation coefficient of the cost function to minimize the cost function.

Here, the learning unit 310 may use an open source library, an engine, and an application programming interface (API). For example, the learning unit 310 may use a library such as 'Spark' or 'mahout', a learning engine such as Google's 'TensorFlow', or an API such as IBM's 'WATSON'.

Here, the learning data of the learning unit 310 may be stored in a separate database (not shown).

Here, the separate database may be a separate cloud server.

[Equation 1]

(Where α is the first correlation coefficient, δ is the second correlation coefficient, M is the number of gas sensors, N is the number of data for learning sensor characteristics according to the gas environment experimental conditions, P is the number of gases to be measured, and S is the gas The measured value of the sensor, C is the partial pressure value of the gas, and l is a natural number from 1 to P)

Here, the first correlation coefficient is a correlation coefficient for a correlation between a partial pressure value of each of the at least one gas and a measured value of the at least one gas sensor, and the second correlation coefficient is a correlation between a partial pressure value of a plurality of gases. It can be a correlation coefficient for.

For example, first, the learning unit 310 in a situation where the type and partial pressure of two gases (first gas (A) and second gas (B)) specified in a mixed gas in a given environment are known, within a given environment. By inputting the measured value measured through the sensor module 100 into the cost function, it is possible to calculate a correlation coefficient of the cost function that minimizes the cost function of Equation 2 below.

[Equation 2]

(Where α is the first correlation coefficient of the first gas, β is the first correlation coefficient of the second gas, δ is the second correlation coefficient, M is the number of gas sensors, and N is the number of data for learning sensor characteristics according to the gas environment experimental conditions. , S is the measured value of the gas sensor, C is the partial pressure value of the gas)

Thereafter, the learning unit 310 may learn this by repeatedly performing a process of calculating a correlation coefficient of a cost function that minimizes a cost function in each environment while changing the number of gas and partial pressure values in a given environment.

On the other hand, as described above, the learning unit 310 may store the correlation coefficients by classifying them into hierarchies or tables in the learned data.

The calculation unit 320 receives the measured value of the gas, the type and number of gases to be measured in the mixed gas, and the partial pressure value of the gas to be measured in the cost function to which the correlation coefficient calculated and learned by the learning unit is applied. Each can be calculated.

For example, the calculator 320 may calculate a partial pressure value of a specific gas to be measured in an actual environment in which the number and partial pressure value of each gas are not known.

Here, the calculator 320 may calculate the partial pressure values of the first gas A and the second gas B, which minimize the cost function, through Equation 3 below.

[Equation 3]

(Where C _A is the partial pressure value of the first gas, C _B is the partial pressure value of the second gas, α is the first correlation coefficient of the first gas, β is the first correlation coefficient of the second gas, δ is the second correlation coefficient, M is Number of gas sensors, S indicates the measured value of the gas sensor)

That is, the calculation unit 320 may generalize a formula for calculating the partial pressure value of the gas, as shown in Equation 4 below, by using the correlation coefficient learned by the learning unit 310, through which the cost function It is possible to calculate a partial pressure value of at least one specified gas to be minimized.

[Equation 4]

(Wherein, C _jk is the target value to be calculated, α is the first correlation coefficient, δ is the second correlation coefficient, M is the number of gas sensors, P is the number of gases to be measured, and S is the measured value of the gas sensor. )

That is, the calculation module 300 inputs a measurement value of at least one gas sensor measured in the mixed gas into a cost function through a learning process, and calculates each partial pressure value of at least one gas that minimizes the cost function can do.

Therefore, the gas measurement system according to the embodiment of the present invention can calculate the partial pressure value that minimizes the overall cost function after learning the optimized relationship coefficient between the gas partial pressure (concentration) and the gas sensor measurement value.

Hereinafter, a gas measurement system according to another embodiment of the present invention will be described with reference to FIG. 3.

Here, the gas measurement system according to another embodiment of the present invention focuses on the protective member for protecting the sensor module.

3 is an exemplary view showing a state in which a sensor module is accommodated in a protective member in a gas measurement system according to another embodiment of the present invention.

Referring to FIG. 3, the gas sensor unit 10 may include a protection member 11 protecting the sensor module 100.

The protection member 11 may be implemented in a form surrounding the sensor module 100, and the outer surface is configured in a spherical shape to easily move to a desired position by an external force such as throwing of a user or the magnetic force of the gas sensor unit 10 You can.

Here, the protective member 11 is preferably made of a synthetic resin having a predetermined elastic force.

A plurality of inlet holes 12 may be formed in the protection member 11 to allow gas molecules to flow into the sensor module 100 accommodated therein.

Further, a sensor module 100 accommodated inside the protection member 11 and a power supply 13 for supplying power for driving the communication module may be disposed inside the protection member 11.

The power supply 13 is preferably composed of a secondary battery cell capable of charging and discharging. Here, although not shown, the power supply 13 includes a power cable for supplying power to the sensor module 100 and the communication module. .

Hereinafter, a gas measurement system according to another embodiment of the present invention will be described with reference to FIG. 4.

Here, the gas measurement system according to another embodiment of the present invention focuses on the renewable energy generation unit.

4 is an exemplary view showing a state in which a renewable energy generating unit is attached to a protection member in a gas measurement system according to another embodiment of the present invention.

Referring to FIG. 4, the gas sensor unit 10 may be provided with at least one renewable energy generator 14 inside the protective member 11 protecting the sensor module 100.

Since the gas sensor unit 10 can be independently input to the gas environment, it is preferable to continuously supply power for driving the sensor module 100 and the communication module accommodated inside the protection member 11.

That is, the renewable energy generation unit 14 may be electrically connected to the sensor module 100 and the communication module.

The renewable energy generation unit 14 may be implemented as a power generation module capable of generating power using renewable energy such as solar power or wind power.

Here, the protective member 11 is preferably transparent so that light can be transmitted and transmitted to the renewable energy generation unit 14.

The above detailed description is to illustrate the present invention. In addition, the above-described content is to describe and describe preferred embodiments of the present invention, and the present invention can be used in various other combinations, modifications and environments. That is, it is possible to change or modify the scope of the concept of the invention disclosed herein, the scope equivalent to the disclosed contents, and / or the scope of the art or knowledge in the art. The embodiments described describe the best state for implementing the technical idea of the present invention, and various changes required in specific application fields and uses of the present invention are possible. Accordingly, the detailed description of the invention is not intended to limit the invention to the disclosed embodiments. In addition, the appended claims should be construed to include other embodiments.

100: sensor module
200: communication module
300: operation module

Claims (7)

In the gas measurement system for measuring the partial pressure value of a specific gas using the measurement value measured by a plurality of gas sensors in the mixed gas,
A calculation module for inputting a measured value of at least one gas sensor measured in the mixed gas into a cost function and calculating each partial pressure value of at least one gas minimizing the cost function; Including,
The calculation module
The measurement function and the partial pressure value measured in a reference environment in which the number and partial pressure value of each gas are known are input to the cost function to calculate a correlation coefficient of the cost function that minimizes the cost function, and the number and partial pressure values of each gas are changed. A learning unit learning this by repeatedly performing a process of calculating a correlation coefficient of a cost function to minimize the cost function; And
Calculation to calculate the partial pressure value of the gas to be measured by inputting the type, number of gas to be measured in the mixed gas and the measured values of a plurality of gas sensors to the cost function applied with the correlation coefficient calculated and learned by the learning unit part; Gas measurement system comprising a.
According to claim 1,
The correlation coefficient includes a first correlation coefficient and a second correlation coefficient,
The first correlation coefficient is a correlation coefficient for a correlation between a partial pressure value of each of the at least one gas and a measured value of the at least one gas sensor,
And the second correlation coefficient is a correlation coefficient for a correlation between partial pressure values of a plurality of gases.
According to claim 2,
The learning unit

To calculate the correlation coefficient of the cost function to minimize the cost function,
Here, α is the first correlation coefficient, δ is the second correlation coefficient, M is the number of gas sensors, N is the number of data for learning sensor characteristics according to the gas environment experimental conditions, P is the number of gases to be measured, S is the gas sensor The measured value of, C is the partial pressure value of the gas, l is a gas measurement system showing any natural number of 1 to P.
According to claim 3,
The calculation unit

To calculate the partial pressure value of at least one gas to minimize the cost function,
Here, C _jk is the target value to be calculated, α is the first correlation coefficient, δ is the second correlation coefficient, M is the number of gas sensors, P is the number of gases to be measured, and S is the gas representing the measured value of the gas sensor. Measuring system.
According to claim 1,
A sensor module including the plurality of gas sensors and
Gas measurement system further comprising a communication module for transmitting the measured value from the sensor module to the calculation module.
The method of claim 5,
Further comprising a protective member surrounding the sensor module,
A gas measurement system in which an inflow hole for introducing the gas is formed in the protection member.
The method of claim 6,
Gas measurement system that is disposed inside the protective member, further comprising a renewable energy generating unit for supplying power to the sensor module.

Images