Risk early warning method based on isocyanate production equipment

Abstract

The invention relates to the technical field of risk early warning of chemical production, and discloses a risk early warning method based on isocyanate production equipment, and the method comprises the steps: carrying out the accelerated degradation test of an electric leakage signal conditioning circuit, and rapidly building a performance degradation model, the reliability of the whole circuit in the production equipment is extrapolated according to the performance degradation model, and whether a response signal is sent out or not is judged according to the change of the reliability, so that the purposes of monitoring and early warning the isocyanate production process are achieved, the safety in the isocyanate production process is improved, potential dangerous situations are found and prevented in time, and the production efficiency is improved. And the probability of accidents is reduced.

Description

Technical Field

[0001] The invention relates to the technical field of risk warning for chemical production, in particular to a risk warning method based on isocyanate production equipment. Background Art

[0002] Isocyanate is a colorless, clear liquid with strong irritation and is soluble in water at low temperatures. Isocyanate is highly corrosive to other materials except stainless steel, nickel, glass and ceramics, especially to iron, steel, zinc, tin, copper or their alloys. In addition, isocyanate is easy to burn when exposed to heat, open flame or oxidant, and releases MIC vapor, nitrogen oxides, carbon monoxide and hydrogen cyanide during combustion. Therefore, isocyanate has a high safety risk during production, storage and transportation, especially in its production process. Once the production equipment fails, it is easy to leak or fire, which will seriously threaten the personal safety of the staff.

[0003] In order to solve the safety problem of chemical production, a patent for a chemical enterprise production safety early warning system is disclosed in the prior art, and the publication number of the patent is: CN114708707A; it uses dust sensors, toxic gas sensors, flammable gas sensors, explosive gas sensors and flame sensors to collect monitoring data of the production environment, and then uses a comprehensive analysis unit to receive the above monitoring data. When an abnormal problem occurs in the environment, that is, when an abnormality occurs in the monitoring data, different analysis and processing can be performed according to different problems. At the same time, the camera image at the problem location can be retrieved to track and trace the problem found.

[0004] However, the above patent mainly uses various sensors to detect changes in the chemical production environment to determine whether a fault has occurred. In actual use, it needs to rely heavily on the performance and accuracy of various sensors. The entire early warning process has great uncertainty and lag, and it is impossible to timely and accurately implement pre-intervention in dangerous situations. In the chemical production process, the operating status of the production equipment is the factor that can most quickly and directly reflect production safety, and the operating status of the production equipment is closely related to the circuit status inside it. Therefore, those skilled in the art need an early warning method based on the state change of isocyanate production equipment. Summary of the invention

[0005] The purpose of the present invention is to solve the above problems and to design a risk early warning method based on isocyanate production equipment.

[0006] To achieve the above-mentioned purpose, the technical solution of the present invention is a risk early warning method based on isocyanate production equipment, which comprises the following steps: Step 1: Set the temperature in the production equipment as the accelerated stress and the operating current value as the degradation characteristic quantity, perform a constant stress accelerated degradation test on the leakage signal conditioning circuit in the production equipment, and obtain degradation data of the leakage signal conditioning circuit; Step 2: According to the degradation data of the leakage signal conditioning circuit and based on the Wiener process, a performance degradation model of the leakage signal conditioning circuit is established, and the unknown parameters of the performance degradation model are obtained by using the maximum likelihood estimation method. Then, the real-time action current value and degradation rate of the circuit of the entire production equipment are substituted into the performance degradation model of the leakage signal conditioning circuit, and then the time when the action current value in the circuit of the entire production equipment exceeds the failure threshold is predicted, and the length of time is regarded as the predicted life of the circuit of the entire production equipment; then, the predicted life distribution of the circuit of the entire production equipment under normal stress is derived by using the life prediction result of the circuit of the entire production equipment in combination with the Arrhenius equation, and then the reliability of the circuit of the entire production equipment is evaluated; Step three: Provide early warning of production equipment failures based on the life prediction results and reliability of the entire production equipment.

[0007] The specific test steps of the accelerated degradation test in step 1 are as follows: 1) The leakage signal conditioning circuit is powered and operated at room temperature for 24 hours, and the voltage of key nodes and the resistance and capacitance of key components are recorded as a benchmark for performance degradation comparison; 2) Accelerated degradation testing of leakage signal conditioning circuits at temperature stress levels exceeding 100°C; 3) Carry out 30 cycles of testing with 24 hours as one test cycle. After each test cycle, cool the circuit to room temperature, measure the performance parameters and continue the next test cycle. The obtained test data is used as the basis for constructing the performance degradation model.

[0008] The construction process of the performance degradation model in step 2 is: The one-dimensional linear Wiener degradation model of the leakage signal conditioning circuit is constructed with the action current value as the degradation characteristic quantity: (3) Where: t is the accelerated degradation test time; λ is the drift parameter; σ is the diffusion parameter; W(t) is the standard Brownian motion; X(t) is the cumulative degradation of the action current value at time t, which has an expectation of λt and a variance of σ 2 Normal distribution of t, that is, X(t)-N(λt,σ 2 t); Let x ef is the segmented degradation of the operating current value of the e-th test circuit in the f-th cycle, where: e=1,2,…,u; j=1,2,…,v; x ef Subject to the expectation of λΔt and the variance of σ2 The normal distribution of Δt, that is, x ef -N(λΔt,σ 2 Δt); use the maximum likelihood estimation method to make point estimates of the drift parameter λ and the diffusion parameter σ, and the corresponding maximum likelihood estimation function is: (4) Taking the logarithm of formula (4) yields: (5) The maximum likelihood estimates obtained by taking partial derivatives of the parameters λ and α in equation (5) are: (6) The life of the leakage signal conditioning circuit is defined as: (7) Where V is the failure threshold of the operating current value of the leakage signal conditioning circuit; From equations (6) and (7), we can get the probability density function of the life T of the leakage signal conditioning circuit: (8) Reliability is defined as the probability that an electronic circuit can complete its set function within a specified time and under specified conditions. (9) Substituting formula (8) into formula (9), the reliability function is obtained as follows: (10) From the above formulas, we can get the expectation and variance of life span T as: (11).

[0009] The reliability evaluation process in step 2 is as follows: Parameters λ and σ are obtained from formula (6), and the relationship between parameters λ, σ and stress level is established. Then, by substituting parameters such as λ and σ into formulas (8) and (10), the life prediction curve of the entire circuit in the production equipment can be obtained, and the reliability can be determined by the life prediction curve.

[0010] The fault warning process of the production equipment in step 3 is as follows: When the service life of the production equipment circuit exceeds the predicted life, the reliability of the production equipment circuit will drop rapidly. After the reliability of the production equipment circuit is lower than the set value, the production equipment will issue a warning message to prompt the staff to perform maintenance on the production equipment.

[0011] The production equipment will display corresponding emergency treatment suggestions while outputting the warning information.

[0012] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention uses the constructed degradation model combined with the statistical method - maximum likelihood estimation to accurately predict the remaining life and reliability level of the generated equipment circuit, with low computational complexity: the existing technology involves multiple algorithms and steps, such as spatial correlation analysis, significance test, hybrid frog leaping mechanism, etc., with high computational complexity and increased node energy consumption; 2. The present invention application is based on mathematical models and algorithms, does not involve methods in the field of artificial intelligence, does not require a large amount of data and computing resources, and does not rely on a large amount of data. However, the prior art requires the application of neural networks for fault diagnosis, which has problems with convergence speed and training, and poor generalization ability on new data. In addition, the effect of the diagnosis method based on artificial neural networks depends largely on the selected neural network structure. Different structures may lead to completely different diagnostic results, which will reduce the accuracy of fault detection; 3. The present invention application does not rely on the huge amount of data generated by the chemical production environment, but uses temperature as the accelerated stress and the action current value as the degradation characteristic quantity to perform fault prediction and reliability evaluation on the leakage signal conditioning circuit of the production equipment. It will not cause low data quality due to sensor failure, affect the diagnosis results, and has strong anti-interference ability. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a flow chart of a risk early warning method based on isocyanate production equipment according to the present invention; Figure 2 is a degradation trend diagram of the action current value of the present invention; Figure 3 is a life prediction curve diagram of the leakage signal conditioning circuit of the present invention; Figure 4 It is the life data and fitting curve diagram of the entire circuit in the production equipment of the present invention; Figure 5 It is the principle diagram of the leakage signal conditioning circuit of the present invention. DETAILED DESCRIPTION

[0014] The present invention will be described in detail below in conjunction with the accompanying drawings. Figure 1-Figure 5 As shown; The degradation test process of the present application uses temperature and humidity as accelerated stress, wherein temperature stress is the main reason affecting the performance degradation of electronic circuits, and humidity stress is a secondary reason. Long-term high temperature will aggravate the degradation of electronic components in production equipment, causing changes in the output performance of the leakage signal conditioning circuit, thereby affecting the control and response of the entire production equipment and causing failures. Therefore, the present application uses the accelerated degradation process of the production equipment circuit under temperature stress as the analysis basis for fault warning, selects the resistance, capacitance and action current values ​​that have a greater impact on the performance of the leakage signal conditioning circuit as measurement variables, and conducts accelerated degradation tests on production equipment; and adds jumper caps at key nodes of the circuit to eliminate measurement errors caused by series and parallel connection of components. After the test cycle is over, the circuit is naturally cooled to room temperature to ensure that the circuit test environment is consistent, and impedance analyzers, multimeters and other equipment are used to measure the resistance, capacitance and other data of key components, while the action current value is measured by indirect measurement, that is, the action current value data is indirectly obtained through the change of the mutual inductor in the current detection circuit.

[0015] The specific test steps of the entire accelerated degradation test are as follows: 1) The leakage signal conditioning circuit is powered and operated at room temperature for 24 hours, and the voltage of key nodes and the resistance and capacitance of key components are recorded as a benchmark for performance degradation comparison; 2) Accelerated degradation testing of leakage signal conditioning circuits at temperature stress levels exceeding 100°C; 3) Carry out 30 cycles of testing with 24 hours as one test cycle. After each test cycle, cool the circuit to room temperature, measure the performance parameters and continue the next test cycle. The obtained test data is used as the basis for constructing the performance degradation model.

[0016] The accelerated degradation test data is processed, and the degradation characteristic quantity of the leakage signal conditioning circuit performance is selected. Figure 2 From the degradation trend diagram of the action current value, it can be seen that the action current value gradually increases with the increase of degradation time, and the higher the temperature, the greater the degradation rate of the action current value. The increase in the action current value will increase the risk of refusal to operate of the leakage circuit breaker. Furthermore, the action current value can reflect the degradation trend of the entire circuit. Therefore, this application selects the action current value as the characteristic quantity of the performance degradation of the leakage signal conditioning circuit, and establishes a univariate linear Wiener degradation model.

[0017] The one-dimensional linear Wiener degradation model of the leakage signal conditioning circuit is constructed with the action current value as the degradation characteristic quantity: (3) Where: t is the accelerated degradation test time; λ is the drift parameter; σ is the diffusion parameter; W(t) is the standard Brownian motion; X(t) is the cumulative degradation of the action current value at time t, which has an expectation of λt and a variance of σ 2 Normal distribution of t, that is, X(t)-N(λt,σ 2 t); Let x ef is the segmented degradation of the operating current value of the e-th test circuit in the f-th cycle, where: e=1,2,…,u; j=1,2,…,v; x ef Subject to the expectation λΔt and variance σ 2 The normal distribution of Δt, that is, x ef -N(λΔt,σ 2 Δt); use the maximum likelihood estimation method to make point estimates of the drift parameter λ and the diffusion parameter σ, and the corresponding maximum likelihood estimation function is: (4) Taking the logarithm of formula (4) yields: (5) The maximum likelihood estimates obtained by taking partial derivatives of the parameters λ and α in equation (5) are: (6) The life of the leakage signal conditioning circuit is defined as: (7) Where V is the failure threshold of the operating current value of the leakage signal conditioning circuit; From equations (6) and (7), we can get the probability density function of the life T of the leakage signal conditioning circuit: (8) Reliability is defined as the probability that an electronic circuit can complete its set function within a specified time and under specified conditions. (9) Substituting formula (8) into formula (9), the reliability function is obtained as follows: (10) From the above formulas, we can get the expectation and variance of life span T as: (11).

[0018] Parameters λ and σ are obtained from formula (6), and the relationship between parameters λ, σ and stress level is established. Substituting parameters such as λ and σ into formulas (8) and (10) can obtain the life prediction curve of the entire circuit in the production equipment, and the reliability is determined by the life prediction curve.

[0019] The basic idea of ​​using accelerated degradation test to predict life is to extrapolate the life of the entire circuit in the production equipment under normal stress by the life of the entire circuit of the production equipment under high temperature stress. The key to the extrapolation process is. The Arrhenius model, that is, the Arrhenius equation, is the most typical and widely used acceleration model, and its expression is: (12) Where: M is the degradation amount of the characteristic quantity of the electronic circuit; k is the Boltzmann constant; T is the absolute temperature; A0 is a constant; t is the degradation time; ΔE is the activation energy of the degradation mechanism; ∂M / ∂t represents the degradation rate at absolute temperature T, which is a linear function of time t.

[0020] (13) Let ΔM=M2-M1, and define the lifetime L as the time difference from t1 to t2 of the electronic circuit, that is, L= t2-t1.

[0021] Taking the logarithm of both sides of equation (13) we get: (14) Let B = ln(ΔM / A0 ), A = ΔE / k, we get: (15) Where: L is the service life of the electronic circuit; T is the absolute temperature; A and B are unknown coefficients, and the logarithm of the life L is linearly related to the inverse of the absolute temperature T.

[0022] The predicted life of the entire circuit of the production equipment is fitted using the least squares method, and the life fitting straight line is obtained as follows: Figure 3 As shown. Figure 3 The slope A, intercept B, sum variance, coefficient of determination and other fitting parameters of the fitting line can be obtained, among which the sum variance SSE = 0.02027, which is very close to 0; the coefficient of determination R-square = 0.9681, which is very close to 1, so it can be considered that the data fitting effect is good, and the equation of the fitting line is: (16) Based on the above formula, it can be predicted that the life of the entire circuit of the production equipment at room temperature is 8-12 years.

[0023] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions. The sentence "includes an element defined by ... does not exclude the existence of other identical elements in the process, method, article or device including the element".

[0024] The above technical solutions only reflect the preferred technical solutions of the technical solutions of the present invention. Some changes that may be made to certain parts thereof by technicians in this technical field all reflect the principles of the present invention and fall within the protection scope of the present invention.

Claims

1. A risk early warning method based on isocyanate production equipment, characterized in that: The method comprises the following steps: Step 1: Set the temperature in the production equipment as the accelerated stress and the operating current value as the degradation characteristic quantity, perform a constant stress accelerated degradation test on the leakage signal conditioning circuit in the production equipment, and obtain degradation data of the leakage signal conditioning circuit; Step 2: According to the degradation data of the leakage signal conditioning circuit and based on the Wiener process, a performance degradation model of the leakage signal conditioning circuit is established, and the unknown parameters of the performance degradation model are obtained by using the maximum likelihood estimation method. Then, the real-time action current value and degradation rate of the circuit of the entire production equipment are substituted into the performance degradation model of the leakage signal conditioning circuit, and then the time when the action current value in the circuit of the entire production equipment exceeds the failure threshold is predicted, and the length of time is regarded as the predicted life of the circuit of the entire production equipment; then, the predicted life distribution of the circuit of the entire production equipment under normal stress is derived by using the life prediction result of the circuit of the entire production equipment in combination with the Arrhenius equation, and then the reliability of the circuit of the entire production equipment is evaluated; Step three: Provide early warning of production equipment failures based on the life prediction results and reliability of the entire production equipment.

2. A risk early warning method based on isocyanate production equipment according to claim 1, characterized in that: The specific test steps of the accelerated degradation test in step 1 are as follows: 1) The leakage signal conditioning circuit is powered and operated at room temperature for 24 hours, and the voltage of key nodes and the resistance and capacitance data of key components are recorded as a benchmark for performance degradation comparison; 2) Accelerated degradation testing of leakage signal conditioning circuits at temperature stress levels exceeding 100°C; 3) Carry out 30 cycles of testing with 24 hours as one test cycle. After each test cycle, cool the circuit to room temperature, measure the performance parameters and continue the next test cycle. The obtained test data is used as the basis for constructing the performance degradation model.

3. A risk early warning method based on isocyanate production equipment according to claim 1, characterized in that: The construction process of the performance degradation model in step 2 is: The one-dimensional linear Wiener degradation model of the leakage signal conditioning circuit is constructed with the action current value as the degradation characteristic quantity: （3） Where: t is the accelerated degradation test time; λ is the drift parameter; σ is the diffusion parameter; W(t) is the standard Brownian motion; X(t) is the cumulative degradation of the action current value at time t, which has an expectation of λt and a variance of σ 2 Normal distribution of t, that is, X(t)-N(λt,σ 2 t); Let x ef is the segmented degradation of the operating current value of the e-th test circuit in the f-th cycle, where: e=1,2,…,u; j=1,2,…,v; x ef Subject to the expectation λΔt and variance σ 2 The normal distribution of Δt, that is, x ef -N(λΔt,σ 2 Δt); use the maximum likelihood estimation method to make point estimates of the drift parameter λ and the diffusion parameter σ, and the corresponding maximum likelihood estimation function is: （4） Taking the logarithm of formula (4) yields: （5） The maximum likelihood estimates obtained by taking partial derivatives of the parameters λ and α in equation (5) are: （6） The life of the leakage signal conditioning circuit is defined as: （7） Where V is the failure threshold of the operating current value of the leakage signal conditioning circuit; From equations (6) and (7), we can get the probability density function of the life T of the leakage signal conditioning circuit: （8） Reliability is defined as the probability that an electronic circuit can complete its set function within a specified time and under specified conditions. （9） Substituting formula (8) into formula (9), the reliability function is obtained as follows: （10） From the above formulas, we can get the expectation and variance of life span T as: （11）。 4. A risk early warning method based on isocyanate production equipment according to claim 3, characterized in that: The reliability evaluation process in step 2 is as follows: Parameters λ and σ are obtained from equation (6), and the relationship between parameters λ, σ and stress level is established. Substituting λ and σ into equations (8) and (10) can obtain the life prediction curve of the circuit, and the reliability is determined by the life prediction curve.

5. A risk early warning method based on isocyanate production equipment according to claim 4, characterized in that: The risk warning process of production equipment in step 3 is as follows: When the service life of the production equipment circuit exceeds the predicted life, the reliability of the production equipment circuit will drop rapidly. After the reliability of the production equipment circuit is lower than the set value, the production equipment will issue a warning message to prompt the staff to perform maintenance on the production equipment.

6. A risk early warning method based on isocyanate production equipment according to claim 5, characterized in that: The production equipment will display corresponding emergency treatment suggestions while outputting the warning information.

Images