Self-adaptive system for multiple gas sensors and parameter calculation method of self-adaptive system

Abstract

The invention discloses a self-adaptive system for various gas sensors and a parameter calculation method of the self-adaptive system, and belongs to the technical field of gas detection. The system comprises a front-end polarity adaptation unit, a current-voltage conversion unit, an intelligent neural network calculation unit U5 and an analog-to-digital conversion unit U4, wherein the front-end polarity adaptation unit is used for being connected with a gas sensor and receiving a current signal of the gas sensor; the signal subjected to front-end polarity matching and current-voltage conversion is subjected to signal filtering and signal gain amplification through the intelligent neural network calculation unit U5, analog quantity and digital quantity conversion is performed through the analog-to-digital conversion unit U4, the detected gas voltage value of the gas sensor is obtained, and the intelligent neural network calculation unit U5 calculates the current gas concentration. The multi-gas detection device adapts to gas sensors of different manufacturers, different principles and different measuring ranges, and intelligent self-adaption of multi-gas detection is achieved on the premise that the product cost is not increased; and accurate concentrations can be calculated for different gases.

Description

Technical Field

[0001] The invention belongs to the technical field of gas detection, and in particular relates to an adaptive system for multiple gas sensors. Background Art

[0002] Gas sensor types include photoionization gas sensors, semiconductor gas sensors, contact combustion gas sensors, electrochemical gas sensors, non-dispersive infrared gas sensors and other types of sensors. The multi-gas detectors currently on the market often adopt a one-time configuration with fixed detection objects. Different types of gas sensors cannot be replaced later, and the detection objects cannot be expanded after leaving the factory. In order to be compatible with more detection objects, the equipment needs to add a lot of sensor hardware and software resources. Some functions are not used on site, resulting in waste of resources and increased costs. These increased costs are ultimately passed on to end customers, causing price increases.

[0003] Most existing technologies increase the number of sensors in the equipment to achieve the purpose of multi-gas detection. Due to factors such as the size and cost of the equipment, most of them adopt semiconductor sensors or MEMS sensor arrays. Due to the defect of gas cross-interference of the sensor, the actual detection data accuracy is not high, which seriously affects the actual use effect. Another technical route is to replace the corresponding sensor for different on-site detection objects to achieve the purpose of multi-gas detection.

[0004] There are two implementation routes in the existing technology:

[0005] Technical route 1: The Chinese utility model patent with publication number CN219391885U discloses a multi-scenario environment adaptive gas semiconductor array gas sensor device using a semiconductor sensor array to deploy multiple sensors at one time to detect multiple gases.

[0006] The problems with the prior art are: placing multiple sensors at one time increases the cost; using semiconductor sensors has poor accuracy and large temperature drift, making it impossible to accurately detect ambient gases, and can only be used as an alarm, with limited application scenarios.

[0007] Technical route 2:

[0008] The Chinese invention patent application with publication number CN118090884A discloses a weak signal measurement system for an adaptive gas sensor, which uses a first-stage transimpedance amplifier circuit, a second-stage amplifier circuit and an analog filter circuit, and uses an MCU unit to control an analog switch to achieve gain control of the second-stage amplifier circuit; the MCU unit reads the digital signal converted by the differential AD converter, and compares it with the internal threshold, controls the analog switch to switch the resistance of the resistor amplification network connected to the second-stage amplifier circuit, and adjusts the amplification factor of the second-stage amplifier circuit. The technical solution of this patent application can only be adapted to different ranges of the same type of sensor, and the gain can be adjusted to meet the amplification of small signals; however, in actual use, due to the different principles of gas sensors, the polarity of the gas sensor output signal has both positive and negative polarity, and the corresponding amplifier circuits are also different; for gas sensors with different polarities, the technical solution of this application cannot adaptively match the polarity of the amplifier circuit. Summary of the invention

[0009] In view of the above-mentioned problems existing in the prior art, the technical problem to be solved by the present invention is to provide an adaptive system for multiple gas sensors and a parameter calculation method thereof, which can adapt to different types of sensors. For different gases to be detected, only the corresponding gas sensors need to be replaced, thereby realizing low-cost intelligent online monitoring of multiple gases.

[0010] Technical solution: In order to solve the above technical problems, the technical solution adopted by the present invention is as follows:

[0011] An adaptive system for multiple gas sensors, including a front-end polarity adapter unit for connecting to a gas sensor and receiving a current signal of the gas sensor, a current-voltage conversion unit connected to the front-end polarity adapter unit, an intelligent neural network calculation unit U5 connected to the current-voltage conversion unit, and an analog-to-digital conversion unit U4 connected to the intelligent neural network calculation unit U5, wherein the intelligent neural network calculation unit U5 realizes signal front-end polarity matching by controlling the front-end polarity adapter unit, and the intelligent neural network calculation unit U5 converts the gas sensor current signal into a voltage signal by controlling the current-voltage conversion unit. The signal after front-end polarity matching and current-voltage conversion is filtered and amplified by the intelligent neural network calculation unit U5, and is sent to the analog-to-digital conversion unit U4 after processing. The intelligent neural network calculation unit U5 controls the analog-to-digital conversion unit U4 to convert analog quantities into digital quantities to obtain the detected gas voltage value of the gas sensor, and the intelligent neural network calculation unit U5 calculates the current gas concentration.

[0012] Preferably, the front-end polarity adaptation unit includes a first polarity conversion unit U2-1 and a second polarity conversion unit U2-2, the current-voltage conversion unit includes a transconductance amplifier U1A, a preamplifier U1B connected to the transconductance amplifier U1A, and a digital resistor U3 connected to the preamplifier U1B, the first polarity conversion unit U2-1 is connected to the transconductance amplifier U1A through a first resistor R1B, the second polarity conversion unit U2-2 is connected to the transconductance amplifier U1A through a second resistor R2B, the intelligent neural network calculation unit U5 adapts the type of the gas sensor, starts switching the first polarity conversion unit U2-1 and the second polarity conversion unit U2-2, controls the polarity matching of the analog switch of the front-end polarity adaptation unit, the gas sensor current signal is connected to the transconductance amplifier U1A through the first polarity conversion unit U2-1 or the second polarity conversion unit U2-2, the transconductance amplifier U1A is connected to the preamplifier U1B through a third resistor R3, and the gas sensor current signal is connected to the transconductance amplifier U1A through the third resistor R4. The three resistors R3 are converted into voltage signals and primary amplified through the preamplifier U1B. The reverse input terminal of the preamplifier U1B is connected to the fourth resistor R4 and connected to the digital resistor U3 through the sixth resistor R6. The intelligent neural network calculation unit U5 automatically calculates the gain of the primary amplification according to the type, range and manufacturer information of the gas sensor, and controls the output resistance of the digital resistor U3. The output terminal of the preamplifier U1B is connected to the intelligent neural network calculation unit U5. The voltage signal output by the preamplifier U1B is digitally filtered and secondary amplified by the intelligent neural network calculation unit U5 to meet the input requirements of the analog-to-digital conversion unit U4. The intelligent neural network calculation unit U5 controls the analog-to-digital conversion unit U4 to convert analog quantities into digital quantities to obtain the detected gas voltage value of the gas sensor. The intelligent neural network calculation unit U5 automatically calculates the sensitivity parameters, sensor environmental calibration parameters and zero point parameters of the gas sensor according to the type, range and manufacturer information of the gas sensor, and finally calculates the current gas concentration.

[0013] Preferably, it also includes a display unit U6, an alarm unit U7 and a remote terminal unit U8 connected to the intelligent neural network computing unit U5, the display unit U6 is used to display the gas detection concentration display and the alarm display, the alarm unit U7 is used for exceeding the standard warning, and the remote terminal unit U8 is used for remote data interaction to obtain the gas sensor type, range and manufacturer information, and synchronize the detected gas concentration and alarm information to the remote data center.

[0014] Preferably, the intelligent neural network calculation unit U5 includes a first part unit and an intelligent parameter calculation unit, wherein the first part unit is used for filtering and amplifying the signal, and the intelligent parameter calculation unit is used for parameter calculation of the gas sensor.

[0015] Preferably, the detected gas passes through the gas sensor, changing the gas concentration into a weak current signal I_Sin, which is output as a positive signal or a negative signal depending on the sensor principle; the I_Sin signal is input into the first polarity conversion unit U2-1 and the second polarity conversion unit U2-2, and the intelligent neural network calculation unit U5 automatically determines the sensor type and controls Sin_CTL1 and Sin_CTL2 to automatically switch the polarity. When I_Sin is a negative signal, the first analog switch S1 of the first polarity conversion unit U2-1 is turned on to connect the signal I_Sin to R1B, and the second analog switch S2 of the second polarity conversion unit U2-2 is turned on; when I_Sin is a positive signal, the first analog switch S1 of the second polarity conversion unit U2-2 is turned on to connect the signal I_Sin to R2B, and S2 of the first polarity conversion unit U2-1 is turned on.

[0016] Preferably, the gas sensor types include photoionization gas sensors, semiconductor gas sensors, contact combustion gas sensors, electrochemical gas sensors and non-dispersive infrared gas sensors, and the measuring range includes 0~10ppm, 10~100ppm, 100~1000ppm, 1000~10000ppm, greater than 10000ppm and 0~100%LEL.

[0017] The present invention also provides a method for calculating parameters of an adaptive system for multiple gas sensors, which is applied to the above-mentioned adaptive system for multiple gas sensors and includes the following steps:

[0018] Step 1: The intelligent neural network calculation unit U5 collects input information, including the type, measuring range and manufacturer information of the gas sensor, and digitizes and collects the input information;

[0019] Step 2: The trained intelligent neural network operation unit U5 outputs the sensor coefficient vector C through forward propagation based on the input information. The sensor coefficient vector C includes the sensor polarity parameter Sin_PN, the sensor first-level gain parameter Ku3, the second-level gain parameter S_AMP, the second-order filter system function parameter, and the concentration zero voltage reference factor AE. 20 , concentration sensitivity parameter M 20 , cross-interference coefficients CF1, CF2…CF n and temperature correction factor rt;

[0020] Step 3: According to the sensor polarity parameter Sin_PN given by the intelligent neural network operation unit U5, Sin_U2-1_CT1 and Sin_U2-2_CT1 are controlled to realize automatic adaptation of the sensor polarity; the current signal I_Sin of the sensor is connected to the U1A circuit to realize the conversion between current and voltage, and the output voltage of the transconductance amplifier U1A is: , where Vref is the zero adjustment voltage, R3 is the resistance value of the third resistor R3, and the output current I_Sin of the gas sensor is converted into a voltage signal;

[0021] Step 4: Based on the primary gain parameter Ku3 of the gas sensor calculated by the intelligent neural network computing unit U5, the intelligent neural network computing unit U5 automatically sets the resistance value of the digital resistor U3. ;

[0022] Step 5: Based on the second-order filter system function parameters and the secondary amplification gain weight parameter S_AMP of the filter of the gas sensor calculated by the intelligent neural network operation unit U5, the intelligent neural network operation unit U5 performs signal filtering and secondary signal amplification processing of the sensor;

[0023] Step 6: The concentration zero voltage reference factor AE of the gas sensor given by the intelligent neural network computing unit U5 20 , concentration sensitivity parameter M 20 , cross-interference coefficients CF1, CF2…CFn and temperature correction factor rt, and the above parameters are used to calculate the sensor concentration.

[0024] Preferably, the resistance value of the digital resistor U3 is set in step 4 , the formula is as follows:

[0025]

[0026] Wherein, Ku3 is the primary gain parameter of the sensor, R4 is the resistance value of the fourth resistor R4, and R6 is the resistance value of the fourth resistor R6.

[0027] Preferably, the system transfer function of the second-order analog filter in step 5 is as follows:

[0028] The corresponding digital analog filter system function:

[0029]

[0030] Where a is the gain coefficient, b is the second-order coefficient, is a complex variable on the unit circle,

[0031] is a natural constant, j is an imaginary unit, and T is the sampling period.

[0032] Preferably, in step 6, interference suppression is required for the cross-interference of the sensors in the ambient air, and the gas concentration calculation formula is as follows:

[0033]

[0034] Among them, AE 20 : The zero output voltage reference factor of the sensor at room temperature of 20°C;

[0035] VE: real-time voltage of the sensor;

[0036] CFn: cross-interference coefficient of the nth gas on the target gas;

[0037] Rair_n: the concentration of the nth gas;

[0038] rt: temperature correction coefficient;

[0039] M 20 : Sensitivity parameters of target gas at room temperature 20℃.

[0040] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0041] 1. Adapt to gas sensors of different manufacturers, different principles and different ranges, and realize intelligent adaptation of multi-gas detection without increasing product costs;

[0042] 2. Through the intelligent neural network calculation unit, the signal polarity of the gas sensor is automatically matched, the range is automatically switched, the analog signal is filtered, and the signal gain is automatically matched, so that the accurate concentration can be calculated for different gases;

[0043] 3. According to the actual detection needs, different types of gas sensors can be connected. A variety of gas sensors can be plug-and-play at the detection site, without increasing hardware costs or replacing instruments and equipment, thus realizing the multi-purpose use of one instrument. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is a system block diagram of an embodiment of the present invention;

[0045] Figure 2 is a schematic diagram of a system circuit of an embodiment of the present invention;

[0046] Figure 3 yes Figure 2 A magnified view of part of the middle and front end polarity adapter unit;

[0047] Figure 4 yes Figure 2A partial enlarged view of the digital resistor U3, the analog-to-digital conversion unit U4 and the intelligent neural network calculation unit U5;

[0048] Figure 5 yes Figure 2 The middle display unit U6 partial area enlarged view;

[0049] Figure 6 is a flow chart of intelligent neural network training in the embodiment;

[0050] Figure 7 Intelligent neural network calculation diagram in the embodiment. DETAILED DESCRIPTION

[0051] The present invention is further illustrated below in conjunction with specific examples. The examples are implemented based on the technical solutions of the present invention. It should be understood that these examples are only used to illustrate the present invention and are not used to limit the scope of the present invention.

[0052] like Figure 1-5 As shown, an adaptive system for multiple gas sensors includes a front-end polarity adaptation unit, a current-voltage conversion unit, an analog-to-digital conversion unit U4, an intelligent neural network calculation unit U5, a display unit U6, an alarm unit U7 and a remote terminal unit U8. The front-end polarity adaptation unit is used to connect to the gas sensor and receive the gas sensor current signal. The front-end polarity adaptation unit includes a first polarity conversion unit U2-1 and a second polarity conversion unit U2-2. The two polarity conversion units use existing digital switches, such as LTC52 series CMOS analog switches. The current-voltage conversion unit includes a transconductance amplifier U1A, a preamplifier U1B and a digital resistor U3. The transconductance amplifier U1A is connected to the front-end polarity adaptation unit. The positive input terminal of the transconductance amplifier U1A is connected to the No. 2 pin of the second polarity conversion unit U2-2 through the second resistor R2B. The reverse input terminal of the transconductance amplifier U1A is connected to the No. 2 pin of the first polarity conversion unit U2-1 through the first resistor R1B. The output terminal of the transconductance amplifier U1A is connected to the positive input terminal of the preamplifier U1B through the third resistor R3. The third resistor R3 is a transconductance resistor. The gas sensor current signal is converted into a voltage signal through the third resistor R3 and is initially amplified through the preamplifier U1B. The reverse input terminal of the preamplifier U1B is connected to the fourth resistor R4, and the reverse input terminal of the preamplifier U1B is also connected to the digital resistor U3 through the sixth resistor R6. The output terminal of the preamplifier U1B is connected to the intelligent neural network computing unit U5. The intelligent neural network computing unit U5 adopts a large-scale logic array FPGA chip. The intelligent neural network computing unit U5 includes a first part unit and an intelligent parameter computing unit. The first part unit ( Figure 2 and Figure 4Part 1) in the figure is a filter and amplifier (Filter+AMP) designed using FPGA, and the intelligent parameter operation unit (AI-Para_CompU) is an intelligent parameter operation unit designed using FPGA. Digital resistor U3 is connected to the I 2 The C bus SCL0 and SDA0 are connected to the intelligent neural network calculation unit U5. The intelligent neural network calculation unit U5 automatically calculates the gain of the primary amplifier according to the type, range and manufacturer information of the gas sensor, controls the output resistance of the digital resistor U3, and the output end of the preamplifier U1B is connected to the first part of the intelligent neural network calculation unit U5. The analog-to-digital conversion unit U4 is connected to the intelligent neural network calculation unit U5. The intelligent neural network calculation unit U5 controls the analog-to-digital conversion unit U4 to convert the analog quantity and the digital quantity to obtain the detected gas voltage value of the gas sensor. The intelligent neural network calculation unit U5 automatically calculates the sensitivity parameter, sensor environmental calibration parameter and zero point parameter of the gas sensor according to the type, range and manufacturer information of the gas sensor, that is, the gas sensor parameter Air_D, and finally calculates the current gas concentration. The display unit U6 is used to display the gas detection concentration display and alarm display, the alarm unit U7 is used for over-standard warning, and the remote terminal unit U8 is used for remote data interaction, obtains the gas sensor type, range and manufacturer information, and synchronizes the detected gas concentration and alarm information to the remote data center.

[0053] The detected gas in the detection site passes through the gas sensor, and the gas concentration changes into a weak current signal I_Sin. The signal is output as a positive signal or a negative signal according to the different sensor principles; the I_Sin signal is input into the first polarity conversion unit U2-1 and the second polarity conversion unit U2-2, and the intelligent neural network calculation unit U5 automatically determines the sensor type and controls Sin_CTL1 and Sin_CTL2 (these two are polarity control signals, Sin_CTL1 is connected to the No. 6 pin of the first polarity conversion unit U2-1, and Sin_CTL2 is connected to the No. 6 pin of the second polarity conversion unit U2-2). The first polarity conversion unit U2-1 and the second polarity conversion unit U2-2 are respectively provided with a first analog switch S1 and a second analog switch S2. When I_Sin is a negative signal, the first analog switch S1 of the first polarity conversion unit U2-1 is turned on, and the signal I_Sin is connected to R1B, and the second analog switch S2 of the second polarity conversion unit U2-2 is turned on; when I_Sin is a positive signal, the first analog switch S1 of the second polarity conversion unit U2-2 is turned on, and the signal I_Sin is connected to R2B, and the S2 of the first polarity conversion unit U2-1 is turned on. I_Sin passes through the transconductance amplifier U1A transconductance operational amplifier, and is converted into a voltage signal through the third resistor R3. After passing through the third resistor R3, it is connected to the preamplifier U1B for primary amplification. The amplification gain = (digital resistor U3+R6) / R4, where the digital resistor U3 is composed of the I of the intelligent neural network calculation unit U5. 2 C bus SCL0, SDA0 control, intelligent neural network computing unit U5 uses large-scale logic circuit FPGA, according to the sensor type, range, manufacturer information and other sensor characteristic information, automatically calculate the primary amplifier gain, through I 2 The C bus controls the output resistance of the digital resistor U3; the voltage signal SOut output by the preamplifier U1B is sent to the first part unit (Part 1) Filter+AMP of the intelligent neural network calculation unit U5 for digital filtering and secondary amplification to meet the input requirements of the analog-to-digital conversion unit U4. The parameters of the Part 1 unit Filter are automatically calculated by the intelligent neural network calculation unit U5 based on the type and range of the gas sensor. The parameters of different gas sensors can be input. The intelligent neural network calculation unit U5 establishes a neural network self-learning knowledge base to automatically match the relevant parameters of the gas sensor.

[0054] The signal Sin_AD after filtering and amplification by the first part unit Filter+AMP of the intelligent neural network calculation unit U5 is connected to the analog-to-digital conversion unit U4; under the timing control of the intelligent neural network calculation unit U5, the analog-to-digital conversion unit U4 converts Sin_AD into a digital quantity VE, and the intelligent neural network calculation unit U5 uses the gas sensor matching parameters in the experience library, the sensor zero-point output voltage reference factor, sensitivity parameters, temperature correction coefficient, cross-interference coefficient, etc., to finally convert Air_D (gas sensor type, range and manufacturer information) into gas concentration; and judges the gas concentration, the intelligent neural network calculation unit U5 controls the alarm unit U7 to automatically warn of exceeding the standard, and the intelligent neural network calculation unit U5 controls the display unit U6 to display the detected concentration and alarm, etc.; the intelligent neural network calculation unit U5 performs remote data interaction through the remote terminal unit U8, obtains the gas sensor parameter Air_D, including the gas sensor type, range, manufacturer information, etc., and synchronizes the detected gas concentration, alarm information, etc. to the remote data center.

[0055] Gas sensor types include photoionization gas sensors (PID), semiconductor gas sensors, contact combustion gas sensors, electrochemical gas sensors and non-dispersive infrared gas sensors (NDIR), etc. Figure 2 and Figure 3 Medium, 3 / 4 Electrode PID NDIR … Sensor represents different types of gas sensors. 3 / 4Electrode means that three-electrode sensor and four-electrode sensor belong to the existing electrochemical gas sensors. PID means photoionization gas sensor. NDIR means non-dispersive infrared gas sensor.

[0056] The measuring range includes 0~10ppm, 10~100ppm, 100~1000ppm, 1000~10000ppm, greater than 10000ppm and 0~100%LEL.

[0057] Manufacturer information includes brand manufacturers such as CITY, Alpha, Honey and SUSA. The manufacturer information represents the principle and polarity requirements of the sensor and can be expanded later.

[0058] The intelligent neural network computing unit U5 is a self-learning network model based on the backpropagation (BP) algorithm. It is designed as a multi-layer feedforward neural network, which is trained by inputting characteristic parameters of sensors and finally outputs the expected output coefficients required for subsequent calculations. Figure 6 As shown, by drawing on the neural network BP algorithm and utilizing the gradient search technology, the intelligent neural network of the intelligent neural network computing unit U5 achieves a self-learning effect through a training process.

[0059] The model of the intelligent neural network computing unit U5 is divided into an input layer, a hidden layer, and an output layer;

[0060] (1) The data of the input layer is the feature vector X=[x1,x2,x3] of the parameter Air_D (gas sensor parameter), and the number of nodes is the number of parameters of Air_D n=3;

[0061] Gas sensor parameters include gas sensor type (principle factor), measurement range (range factor) and manufacturer information (brand factor);

[0062] Gas sensor types include photoionization gas sensor (PID), semiconductor gas sensor, contact combustion gas sensor, electrochemical gas sensor and non-dispersive infrared gas sensor (NDIR), etc.

[0063] The measuring range includes 0~10ppm, 10~100ppm, 100~1000ppm, 1000~10000ppm, greater than 10000ppm and 0-100%LEL;

[0064] Manufacturer information includes brand manufacturers such as CITY, Alpha, SUSA, Honey, etc. The manufacturer information represents the principle and polarity requirements of the sensor.

[0065] (2) The number of hidden layers is set to 6, which is set according to the actual complexity of the task to avoid overfitting. The number of nodes in this layer, i, is set to twice the number of input parameters, that is, i=6, and the Softmax function is used as the activation function.

[0066] (3) The number of nodes m in the output layer is consistent with the dimension of the subsequent coefficient calculation, and the coefficient vector C = [c1, c2, …c m ].

[0067] The subsequent calculation coefficient vector C of the sensor includes: sensor polarity parameter, sensor primary gain parameter, secondary gain parameter, filter coefficient, concentration zero voltage reference factor, concentration sensitivity parameter, cross interference coefficient and temperature correction coefficient.

[0068] The BP learning algorithm process is as follows:

[0069] Step 1: Forward propagation

[0070] Step 1.1, the input layer receives the gas sensor feature vector X;

[0071] Step 1.2, hidden layer calculates activation value layer by layer:

[0072] h k = f(W k ﹒hk-1 +b k )

[0073] Among them, h k-1 is the activation value of each layer, W k is the weight matrix, b k is the bias, f is the activation function;

[0074] Step 1.3, obtain the weights and biases from the hidden layer to the output layer, and generate the coefficient vector C of the output layer;

[0075] Step 2: Backward Propagation

[0076] Step 2.1. Loss function: mean square error (MSE) or other task-related loss:

[0077]

[0078] in, is the true coefficient label, Generate coefficients for the output layer, m is the number of output coefficients, is the mean square error;

[0079] Step 2.2, gradient calculation:

[0080] Calculate the weight gradient of each layer by chain rule and bias gradient ;

[0081] Update the parameters using gradient descent:

[0082]

[0083] in, is the learning rate, is the weight gradient, is the bias gradient, b k For bias.

[0084] The self-learning mechanism process is as follows:

[0085] (1)Online learning

[0086] Real-time update: During the deployment phase, the newly collected sensor parameters Air_D and the corresponding true coefficient labels are updated. , continuously update the network parameters, including the activation values ​​of the hidden layers and the generation coefficients of the output layer.

[0087] Incremental training: Only 1-2 new data are used each time to avoid global retraining.

[0088] (2) Adaptive adjustment

[0089] Dynamic learning rate: Use Adam optimizer to automatically adjust the learning rate;

[0090] Regularization: Add Dropout strategy to prevent overfitting.

[0091] The trained intelligent neural network computing unit U5 obtains the best weight gradient of each layer , bias gradient , the intelligent neural network calculation unit U5 can calculate the subsequent calculation coefficient vector C of the sensor based on the input sensor characteristic parameters, which is used for automatic calculation of the gas sensor parameters.

[0092] like Figure 7 As shown, this embodiment also discloses a method for calculating adaptive parameters of multiple gas sensors, which is applied to the above-mentioned adaptive system for multiple gas sensors, and includes the following steps:

[0093] Step 1: Data preprocessing. The intelligent neural network computing unit U5 collects the input Air_D (gas sensor parameters) of the sensor to be replaced, and the information input is sent to the intelligent neural network computing unit U5 through the serial port. The input information includes the gas sensor type, range and manufacturer information, and the input information is digitized and collected; the text information is converted into digital quantities, and data collection processing is performed to condense the input feature parameters into numbers [0, 1, 2, 3, 4, 5] to facilitate the input of subsequent algorithms; and the training set, validation set, and test set are divided multiple times. The corresponding numbers of the sensor parameters are as follows:

[0094] (1) Gas sensor type (Air_D.Type):

[0095] 0: Photoionization gas sensor; 1: Non-dispersive infrared gas sensor; 2: Electrochemical gas sensor; 3: Semiconductor; 4: Contact combustion gas sensor;

[0096] (2) Measuring range (Air_D.Range):

[0097] 0: 0~10ppm; 1: 10~100ppm; 2: 100~1000ppm; 3: 1000~10000ppm;

[0098] 4: greater than 10000ppm; 5: 0-100%LEL;

[0099] (3) Manufacturer information (Air_D.Factory): Four brand manufacturers: CITY, Alpha, Honey and SUSA. The aggregation method is to assign digital quantities: CITY: 0; Alpha: 1; Honey: 2; SUSA: 3.

[0100] Step 2: During the training phase, the intelligent neural network computing unit U5 assigns the digital sensor characteristic parameters calculated in step 1 to the sensor characteristic vector X, and inputs the characteristic vector X into the intelligent neural network U5 through the communication port to obtain the network prediction coefficient. , perform algorithm training:

[0101] Step 2.1: Input sensor feature vector X → network prediction coefficient → Calculate the loss L → Back propagate to update the parameters W k .

[0102] Step 2.2: Stop training based on the validation set loss.

[0103] Step 3: In the inference phase, input the feature vector X of the new sensor parameters, and output the sensor coefficient vector C through forward propagation, including the sensor polarity parameter Sin_PN, the sensor first-order gain parameter Ku3, the second-order gain parameter S_AMP, the second-order filter system function parameters (a, b), and the concentration zero point AE. 20 , concentration sensitivity parameter M 20 , cross interference coefficient CF n And the temperature correction factor rt.

[0104] Step 4: Sensor polarity change unit settings

[0105] According to the sensor polarity parameter Sin_PN given by the intelligent neural network computing unit U5, Sin_U2-1_CT1 and Sin_U2-2_CT1 are controlled to realize automatic adaptation of the sensor polarity. The current signal I_Sin of the sensor is connected to the U1A circuit to realize the conversion between current and voltage. The output voltage of U1A is: , wherein Vref is the zero adjustment voltage, R3 is the resistance value of the third resistor R3, and the third resistor R3 converts the output current I_Sin of the gas sensor into a voltage signal.

[0106] Step 5: Based on the primary gain parameter Ku3 of the gas sensor calculated by the intelligent neural network computing unit U5, the intelligent neural network computing unit U5 automatically sets the resistance value of the digital resistor U3. , the formula is as follows:

[0107]

[0108] Wherein, R4 is the resistance value of the fourth resistor R4, and R6 is the resistance value of the fourth resistor R6;

[0109] Step 6: Based on the second-order filter system function parameters (including gain coefficient a and second-order term coefficient b) and secondary amplification gain weight parameter S_AMP of the gas sensor filter Filter calculated by the intelligent neural network calculation unit U5, the intelligent neural network calculation unit U5 automatically performs signal filtering and secondary signal amplification processing of the sensor;

[0110] The system transfer function of the second-order analog filter is as follows:

[0111]

[0112] In the formula, is the transfer function, is a complex variable in the complex plane, a is the gain coefficient, and b is the second-order term coefficient;

[0113] The corresponding digital analog filter system function:

[0114]

[0115] In the formula, is a system function, is a complex variable on the unit circle, is a natural constant, j is an imaginary unit, a is a gain coefficient, b is a second-order term coefficient, and T is a sampling period.

[0116] Step 7: The concentration zero voltage reference factor AE of the gas sensor is calculated by the intelligent neural network computing unit U5. 20 , sensitivity parameter M 20 , cross-interference coefficients CF1, CF2…CFn, and temperature correction coefficient rt, the intelligent neural network calculation unit U5 uses the above parameters for calculating the sensor concentration;

[0117] In view of the cross-interference of sensors in the ambient air, interference suppression is required. The gas concentration calculation formula is as follows:

[0118]

[0119] Among them, AE 20 : The zero output voltage reference factor of the sensor at room temperature of 20°C;

[0120] VE: real-time voltage of the sensor;

[0121] CFn: cross-interference coefficient of the nth gas on the target gas;

[0122] Rair_n: the concentration of the nth gas (data acquired in advance);

[0123] rt: temperature correction coefficient;

[0124] M 20 : Sensitivity parameters of target gas at room temperature 20℃.

[0125] Take a gas sensor as an example:

[0126] Honeywell's H2S sensor has a range of 0-10ppm and is an electrochemical sensor. It is connected to the system of this embodiment. The manufacturer information, range, and sensor type are input into the device through the serial port. The intelligent neural network computing unit U5 calculates the sensor's polarity, gain, filter coefficient, concentration calculation coefficient and other parameters. Finally, the corresponding gas concentration is obtained through the sensor current value.

[0127] Step 1: Extract the characteristic parameter Air_D of the sensor through data preprocessing:

[0128] Air_D={2,0,2}; / / Air_D.Ttype=2; Air_D.Range=0; Air_D.Factory=2.

[0129] Step 2: Training phase (preliminary stage completed).

[0130] Step 3: Inference Phase

[0131] Input sensor parameter feature vector X (X0=Air_D.Ttype, X1= Air_D.Range, X2= Air_D.Factory):

[0132] X={2;0;2};

[0133] Get the sensor coefficient vector C={

[0134] Sin_PN={ Sin_U2-1_CT1=1, Sin_U2-2_CT1=0}; / / Positive polarity signal

[0135] Ku3=20;

[0136] S_AMP=50;

[0137] a={1,-1.1429,0.4128};

[0138] b={0.0674, 0.1349, 0.06745};

[0139] AE 20 =250;

[0140] M 20 =0.9;

[0141] CF={0.0005,0.2}; / / Interference coefficient of H2S CO, SO2

[0142] rt = 0.9; / / 0.85-1.10

[0143] }.

[0144] Step 4: Set up the sensor polarity change unit. The output of the sensor is a positive polarity signal, and the positive terminal of U1A should be used for input.

[0145] Sin_U2-1_CT1=1;

[0146] Sin_U2-2_CT1=0.

[0147] Step 5: Set the resistance value of digital potentiometer U3 ;

[0148] According to the formula,

[0149]

[0150] In the formula, ,

[0151] .

[0152] Step 6: Set the digital filter and signal secondary gain:

[0153] Set the filter according to parameters a and b;

[0154] Set the secondary gain coefficient of the signal according to S_AMP=5.

[0155] Step 7:

[0156] Use 1ppm standard gas to pass through the H2S sensor, at this time the output current of the sensor I_Sin=0.85uA;

[0157] Resistor ,

[0158]

[0159] Then VE=433.mV;

[0160] Concentration of cross-interference gas CO Other sensors have measured data); concentration of cross-interference gas SO2 (Data measured by other sensors);

[0161] Substitute the data into the following formula:

[0162]

[0163] M_Con(ppm)=1.025ppm.

[0164] The error with theoretical data is 2.5%, which meets the design requirement of ±10%.

[0165] The above are only preferred embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. An adaptive system for multiple gas sensors, characterized in that: It includes a front-end polarity adapter unit for connecting to a gas sensor and receiving a current signal from the gas sensor, a current-voltage conversion unit connected to the front-end polarity adapter unit, an intelligent neural network calculation unit U5 connected to the current-voltage conversion unit, and an analog-to-digital conversion unit U4 connected to the intelligent neural network calculation unit U5. The intelligent neural network calculation unit U5 realizes signal front-end polarity matching by controlling the front-end polarity adapter unit. The intelligent neural network calculation unit U5 converts the gas sensor current signal into a voltage signal by controlling the current-voltage conversion unit. The signal after front-end polarity matching and current-voltage conversion is filtered and amplified by the intelligent neural network calculation unit U5, and is sent to the analog-to-digital conversion unit U4 after processing. The intelligent neural network calculation unit U5 controls the analog-to-digital conversion unit U4 to convert analog quantities into digital quantities to obtain the detected gas voltage value of the gas sensor. The intelligent neural network calculation unit U5 calculates the current gas concentration.

2. The adaptive system for multiple gas sensors according to claim 1, characterized in that: The front-end polarity adaptation unit includes a first polarity conversion unit U2-1 and a second polarity conversion unit U2-2, the current-voltage conversion unit includes a transconductance amplifier U1A, a preamplifier U1B connected to the transconductance amplifier U1A, and a digital resistor U3 connected to the preamplifier U1B, the first polarity conversion unit U2-1 is connected to the transconductance amplifier U1A through a first resistor R1B, the second polarity conversion unit U2-2 is connected to the transconductance amplifier U1A through a second resistor R2B, the intelligent neural network calculation unit U5 adapts the type of the gas sensor, starts switching the first polarity conversion unit U2-1 and the second polarity conversion unit U2-2, controls the polarity matching of the analog switch of the front-end polarity adaptation unit, the gas sensor current signal is connected to the transconductance amplifier U1A through the first polarity conversion unit U2-1 or the second polarity conversion unit U2-2, the transconductance amplifier U1A is connected to the preamplifier U1B through a third resistor R3, the gas sensor current signal is connected to the transconductance amplifier U1A through the third resistor R4, and the gas sensor current signal is connected to the transconductance amplifier U1A through the third resistor R5. The resistor R3 is converted into a voltage signal and is primarily amplified by the preamplifier U1B. The reverse input terminal of the preamplifier U1B is connected to the fourth resistor R4 and is connected to the digital resistor U3 through the sixth resistor R6. The intelligent neural network calculation unit U5 automatically calculates the gain of the primary amplification according to the type, range and manufacturer information of the gas sensor, and controls the output resistance of the digital resistor U3. The output terminal of the preamplifier U1B is connected to the intelligent neural network calculation unit U5. The voltage signal output by the preamplifier U1B is digitally filtered and secondary amplified by the intelligent neural network calculation unit U5 to meet the input requirements of the analog-to-digital conversion unit U4. The intelligent neural network calculation unit U5 controls the analog-to-digital conversion unit U4 to convert analog quantities into digital quantities to obtain the detected gas voltage value of the gas sensor. The intelligent neural network calculation unit U5 automatically calculates the sensitivity parameters, sensor environmental calibration parameters and zero point parameters of the gas sensor according to the type, range and manufacturer information of the gas sensor, and finally calculates the current gas concentration.

3. The adaptive system for multiple gas sensors according to claim 2, characterized in that: It also includes a display unit U6, an alarm unit U7 and a remote terminal unit U8 connected to the intelligent neural network computing unit U5. The display unit U6 is used to display the gas detection concentration and alarm display, the alarm unit U7 is used for exceeding the standard warning, and the remote terminal unit U8 is used for remote data interaction to obtain the gas sensor type, range and manufacturer information, and synchronize the detected gas concentration and alarm information to the remote data center.

4. The adaptive system for multiple gas sensors according to claim 2, characterized in that: The intelligent neural network calculation unit U5 includes a first part unit and an intelligent parameter calculation unit. The first part unit is used for filtering and amplifying signals, and the intelligent parameter calculation unit is used for parameter calculation of the gas sensor.

5. The adaptive system for multiple gas sensors according to claim 2, characterized in that: The detected gas passes through the gas sensor, changing the gas concentration into a weak current signal I_Sin, which is output as a positive signal or a negative signal depending on the sensor principle; the I_Sin signal is input into the first polarity conversion unit U2-1 and the second polarity conversion unit U2-2, and the intelligent neural network calculation unit U5 automatically determines the sensor type and controls Sin_CTL1 and Sin_CTL2 to automatically switch polarity. When I_Sin is a negative signal, the first analog switch S1 of the first polarity conversion unit U2-1 is turned on to connect the signal I_Sin to R1B, and the second analog switch S2 of the second polarity conversion unit U2-2 is turned on; when I_Sin is a positive signal, the first analog switch S1 of the second polarity conversion unit U2-2 is turned on to connect the signal I_Sin to R2B, and S2 of the first polarity conversion unit U2-1 is turned on.

6. The adaptive system for multiple gas sensors according to claim 1, characterized in that: The gas sensor types include photoionization gas sensors, semiconductor gas sensors, contact combustion gas sensors, electrochemical gas sensors and non-dispersive infrared gas sensors, and the measurement ranges include 0~10ppm, 10~100ppm, 100~1000ppm, 1000~10000ppm, greater than 10000ppm and 0~100%LEL.

7. A method for calculating parameters of an adaptive system for multiple gas sensors, applied to an adaptive system for multiple gas sensors as claimed in any one of claims 2 to 6, characterized in that: The following steps are involved: Step 1: The intelligent neural network calculation unit U5 collects input information, including the type, measuring range and manufacturer information of the gas sensor, and digitizes and collects the input information; Step 2: The trained intelligent neural network operation unit U5 outputs the sensor coefficient vector C through forward propagation based on the input information. The sensor coefficient vector C includes the sensor polarity parameter Sin_PN, the sensor first-level gain parameter Ku3, the second-level gain parameter S_AMP, the second-order filter system function parameter, and the concentration zero voltage reference factor AE. 20 , concentration sensitivity parameter M 20 , cross-interference coefficients CF1, CF2…CF n and temperature correction factor rt; Step 3: According to the sensor polarity parameter Sin_PN given by the intelligent neural network operation unit U5, Sin_U2-1_CT1 and Sin_U2-2_CT1 are controlled to realize automatic adaptation of the sensor polarity; the current signal I_Sin of the sensor is connected to the U1A circuit to realize the conversion between current and voltage, and the output voltage of the transconductance amplifier U1A is: , where Vref is the zero adjustment voltage, R3 is the resistance value of the third resistor R3, and the output current I_Sin of the gas sensor is converted into a voltage signal; Step 4: Based on the primary gain parameter Ku3 of the gas sensor calculated by the intelligent neural network computing unit U5, the intelligent neural network computing unit U5 automatically sets the resistance value of the digital resistor U3. ; Step 5: Based on the second-order filter system function parameters and the secondary amplification gain weight parameter S_AMP of the filter of the gas sensor calculated by the intelligent neural network operation unit U5, the intelligent neural network operation unit U5 performs signal filtering and secondary signal amplification processing of the sensor; Step 6: The concentration zero voltage reference factor AE of the gas sensor given by the intelligent neural network computing unit U5 20 , concentration sensitivity parameter M 20 , cross-interference coefficients CF1, CF2…CFn and temperature correction factor rt, and the above parameters are used to calculate the sensor concentration.

8. The adaptive system parameter calculation method for multiple gas sensors according to claim 7, characterized in that: In step 4, the resistance value of the digital resistor U3 is set , the formula is as follows: ； Wherein, Ku3 is the primary gain parameter of the sensor, R4 is the resistance value of the fourth resistor R4, and R6 is the resistance value of the fourth resistor R6.

9. The adaptive system parameter calculation method for multiple gas sensors according to claim 7, characterized in that: The system transfer function of the second-order analog filter in step 5 is as follows: ; The corresponding digital analog filter system function: ； Where a is the gain coefficient, b is the second-order coefficient, is a complex variable on the unit circle, is a natural constant, j is an imaginary unit, and T is the sampling period.

10. The adaptive system parameter calculation method for multiple gas sensors according to claim 7, characterized in that: In step 6, interference suppression is required for the sensor cross-interference in the ambient air. The gas concentration calculation formula is as follows: ； Among them, AE 20 : The zero output voltage reference factor of the sensor at room temperature of 20°C; VE: real-time voltage of the sensor; CFn: cross-interference coefficient of the nth gas on the target gas; Rair_n: the concentration of the nth gas; rt: temperature correction coefficient; M 20 : Sensitivity parameters of target gas at room temperature 20℃.

Images