Reliability evaluation model construction method for cigarette quality supervision and measurement system
By using the PHM failure rate model built with two-parameter Weibull distribution and exponential distribution in the cigarette quality supervision and measurement system, the problem of equipment failure rate and life distribution evaluation is solved, the system reliability and production efficiency are improved, and the early stage of equipment failure is realized. Predictive and preventive maintenance.
Patent Information
- Application Number
- CN202410900887.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-05
- Publication Date
- 2024-12-06
AI Technical Summary
Equipment reliability assessment in cigarette quality supervision and measurement systems is difficult to accurately predict failure rate and life distribution, affecting production efficiency and equipment maintenance management.
Using two-parameter Wimble distribution as the benchmark function and exponential distribution as the connection function, the PHM device failure rate model is constructed. By analyzing the equipment's historical fault data and current state data, the equipment's failure rate and life distribution are estimated, and the state transition probability matrix is established. .
It improves the reliability and production efficiency of the cigarette quality supervision and measurement system, reduces equipment downtime and maintenance costs, and realizes preliminary prediction and preventive maintenance of equipment failures.
Smart Images

Figure CN119089103A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of tobacco quality monitoring, and in particular relates to a reliability evaluation model construction method for a cigarette quality supervision and measurement system. Background Art
[0002] In order to ensure the continuous improvement of cigarette quality level, meet consumer demand, and align with international standards, a new series of national standards for cigarettes have been officially implemented. The new national standards have made major revisions and additions to some technologies, indicators and test methods in the original national standards. In order to fully implement the new national standards and improve the efficiency of quality inspection agencies at all levels, it is necessary to standardize the data management of industry quality inspection departments to ensure the consistency and authenticity of data in industry quality supervision and inspection departments at all levels. The measurement process control is the control of the credibility of quality data and the technical guarantee of quality control. Only when the data is credible can the quality be reliable. Therefore, the reliability evaluation of tobacco quality supervision and measurement instruments is the basis for ensuring the real-time and effective operation of the tobacco quality supervision system.
[0003] The tobacco industry and its related production and operation enterprises have been conducting more extensive explorations on how to manage and improve the quality of tobacco products and improve the quality of tobacco product standards. Among them, in line with market and consumer demand, strengthening the management and control of the product quality standard system has become a common rational choice for the industry and enterprises.
[0004] Cigarette quality monitoring and measuring instruments are usually used to monitor and test various quality indicators in the cigarette production process to ensure that the quality of cigarette products meets the standard requirements. These instruments can include the following types: cigarette appearance detection instruments, tobacco moisture detection instruments, tobacco moisture content detection instruments, smoke component analysis instruments, and tar content detection instruments.
[0005] The Bayesian network model is a reliability analysis method based on a probabilistic graphical model. By constructing a causal relationship network between equipment status and faults, the Bayesian theorem can be used to make inferences and evaluate the reliability of equipment. In the prior art, fuzzy membership functions are used to discretize continuous variables, and expert knowledge is used to construct a Bayesian network to represent conditional probabilities. Through simulation verification, two typical spatial scenarios of approach and collision were processed. The results showed that the model can accurately characterize the threat level of spatial targets, has practical significance, and can handle multivariate relationships.
[0006] The Markov chain model is a random process model commonly used to describe the state transition of equipment. By establishing the transition probability matrix of the equipment between different states, the distribution of the equipment in each state can be predicted, thereby evaluating its reliability. The prior art proposes a strategy to solve the problems of low efficiency and poor environmental adaptability of automatic assembly of industrial robots. First, by establishing the geometric and mechanical models of the shaft and hole under various contact states, the geometric constraints and force characteristics in the parts assembly process are analyzed to obtain the assembly trajectory. Secondly, a few-sample teaching and learning method based on hidden Markov chain model and Gaussian mixture regression is adopted to obtain the expected contact force in the assembly process, and the theoretical trajectory is compensated by the admittance controller to ensure the accuracy and robustness of the contact motion in the actual assembly process.
[0007] PHM technology is a comprehensive health management technology that monitors the operating status of equipment, predicts its remaining life and failure probability, and optimizes maintenance decisions. PHM technology is often combined with machine learning and data mining technology, using a large amount of historical data for model training and fault prediction. For example, a study used PHM technology to predict aircraft engine faults, achieving fault warning and maintenance optimization.
[0008] With the development of Industry 4.0 and Internet of Things technology, the application of big data in equipment reliability assessment is becoming more and more extensive. By analyzing equipment operation data, maintenance records and environmental data, equipment failure modes and potential risks can be discovered. By introducing fault prediction and health management (PHM) technology, using the on-board equipment data and image information transmitted back in real time by EOAS, through general fault modeling, big data technology and image recognition, intelligent analysis of equipment operation status changes, online real-time monitoring, fault alarm and prediction, health information management, providing accurate and preventive maintenance, thereby transforming fault handling into fault prevention management. The system reduces the requirements for maintenance personnel, shortens the detection time, effectively improves the efficiency and quality of equipment maintenance, and ensures the reliable operation of on-board equipment.
[0009] In the cigarette industry, with the continuous advancement of cigarette production technology, higher requirements are placed on the reliability and maintenance management of production equipment. Summary of the invention
[0010] The technical problem solved by the present invention is: to provide a reliability evaluation model construction method for a cigarette quality supervision and measurement system, which estimates the failure rate and life distribution of the equipment by analyzing the historical failure data and current status data of the equipment, and proposes a corresponding optimization strategy to improve the reliability and production efficiency of the system.
[0011] Technical solution: In order to solve the above technical problems, the technical solution adopted by the present invention is as follows:
[0012] A reliability evaluation model construction method for a cigarette quality supervision and measurement system includes using a two-parameter Weibull distribution as a benchmark function and an exponential distribution as a link function to establish a PHM equipment failure rate model, and estimating the equipment failure rate and life distribution by analyzing the equipment's historical failure data and current status data.
[0013] Furthermore, the specific forms of the benchmark function and the connection function are as follows:
[0014] h(t,Z(t))=h0(t)ψ(Z(t)),t≥0
[0015] Where: t is the total operating time of the equipment after it is put into operation; Z(t) is the state of the equipment at time t; the benchmark failure rate function h0(t) is used to represent the aging process of the equipment, and the Weibull distribution is selected; the connection function ψ(Z(t)) reflects the impact of different states Z(t) on the equipment failure rate.
[0016] 3. The reliability evaluation model construction method for the cigarette quality monitoring and measurement system according to claim 1 is characterized in that: Weibull distribution is a statistical distribution used to describe the life of equipment, and its probability density function is:
[0017]
[0018] Where x is the equipment failure time, β is the shape parameter, and η is the scale parameter. Based on the Weibull distribution, the probability of equipment failure within a given time is calculated, which is recorded as the failure rate.
[0019] Furthermore, the probability of a device failing within time tt can be expressed by the cumulative distribution function F(t):
[0020]
[0021] The probability of a device failing in the time interval [t, t+Δt][t, t+Δt] can be calculated using the following formula:
[0022] P(t,Δt)=F(t+Δt)-F(t).
[0023] Furthermore, for a new device, t0=0, the device state at the current moment is Z(0)=1, and the reliability function of the device is:
[0024]
[0025] Where T represents the trouble-free operation time of the equipment, h(s,Zj): represents that at time s, the equipment is in state Z jThe hazard rate function h reflects the instantaneous rate of failure of the equipment under a specific state and time.
[0026] For an old device, the device has been running for t0 time, and t0≠0. Assuming that the current device state is Z(t0)=j, and the device state at time t is Z(t)=k, the reliability of the device at time t is:
[0027]
[0028] Among them, X k represents the time that the device stays in state k, assuming that the length of the stay time is random; T k represents the transition time of the kth state, Z j represents the state of the device at time t0, and ZK represents the state of the device at time t.
[0029] Furthermore, after the data model is established, data generation and analysis are performed. The state transition probability matrix is obtained by analyzing the historical state transition data of the equipment, recording the operation of the equipment in different states, and then counting the number of times the equipment transfers to other states in each state, and normalizing it to obtain the transition probability. The above transition probabilities constitute the state transition probability matrix.
[0030] Furthermore, each element Pi of the transition probability matrix P j The probability of transitioning from state i to state j is calculated by dividing the number of transitions by the total number of transitions to the corresponding state. j The calculation formula is as follows:
[0031]
[0032] Where n represents the total number of states, Represents the total number of transitions of state i. According to this formula, each element of the transition probability matrix P is calculated in turn.
[0033] Furthermore, the equipment reliability model adopts the proportional failure model, the benchmark function uses the two-parameter Weibull distribution, and the link function uses the exponential distribution. The parameters of the model are obtained by evaluating historical data. The benchmark function h0(s) and link function ψ(Z(s)) in the proportional probability model are as follows:
[0034]
[0035] ψ(Z(s)))=e 0.5(Z(s)-1)
[0036] Among them, Z(s) represents the aging status of the equipment.
[0037] Beneficial effects: Compared with the prior art, the present invention has the following advantages:
[0038] The reliability evaluation model construction method for the cigarette quality supervision and measurement system of the present invention is based on the proportional failure model (PHM), adopts the two-parameter Weibull distribution as the benchmark function, and the exponential distribution as the connection function to construct a reliability evaluation model for the cigarette quality supervision and measurement system. By analyzing the historical failure data and current state data of the equipment, the failure rate and life distribution of the equipment are estimated, and the corresponding optimization strategy is proposed to improve the reliability and production efficiency of the system.
[0039] The equipment failure rate model of the cigarette quality monitoring and measuring instrument PHM is designed to improve the reliability and efficiency of equipment on the cigarette production line. The model of the present invention combines the characteristics of cigarette production equipment and PHM technology to predict equipment failure in advance, thereby reducing downtime and maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] Figure 1 It is a schematic diagram of the aging failure process under the continuous transfer of equipment status. DETAILED DESCRIPTION
[0041] 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.
[0042] There are currently two main methods used in the research on equipment availability assessment: one is to use statistical data for modeling, which is simple but cannot consider the impact of the current status of the equipment on reliability; the other is to comprehensively consider the health status of the equipment, combining historical data and current status to obtain reliability indicators such as reliability and remaining life.
[0043] PHM-based equipment failure rate model, PHM (Prognostics and Health Management) is a predictive maintenance and health management technology that usually combines historical data, health monitoring data and prediction algorithms of equipment to predict the future failure rate of equipment, so as to predict equipment failure in advance. The equipment failure rate model and related parameters are obtained by using the health status of the comprehensive equipment during operation, based on historical data evaluation, and the reliability indicators of the equipment, such as reliability and remaining life, are obtained by combining the current status of the equipment. This proportional probability model is common in reliability engineering and can be used to evaluate the failure probability and life distribution of equipment.
[0044] Specifically, this model may include the following aspects:
[0045] (1) Data collection and monitoring: A standardized process is used to collect the working status data of cigarette production equipment and the data is averaged.
[0046] (2) Model establishment: Use statistical analysis methods to establish a prediction model for equipment failure rate. This model will take into account the working conditions per unit time and predict the equipment failure rate in the future.
[0047] (3) Failure rate prediction: Based on the established model, the failure rate of the equipment is predicted. This can be a long-term trend prediction or a short-term failure probability prediction, helping production line managers make corresponding maintenance plans and decisions.
[0048] (4) Application of prediction results: Based on the predicted failure rate results, maintenance plans and preventive maintenance strategies are formulated to timely dispatch maintenance personnel and resources, reduce equipment downtime, and improve production efficiency and product quality.
[0049] The reliability evaluation model construction method for the cigarette quality supervision and measurement system of the present invention adopts the two-parameter Weibull distribution as the reference function and the exponential distribution as the connection function to establish the PHM equipment failure rate model. By analyzing the historical failure data and current status data of the equipment, the failure rate and life distribution of the equipment are estimated. The parameters of the PHM equipment failure rate model can be obtained by evaluating the historical data.
[0050] The specific forms of the benchmark function and the connection function are as follows:
[0051] h(t,Z(t))=h0(t)ψ(Z(t)),t≥0
[0052] Where: t is the total operating time of the equipment after it is put into operation; Z(t) is the state of the equipment at time t; the benchmark failure rate function h0(t) is used to represent the aging process of the equipment, and Weibull distribution is often used; the connection function ψ(Z(t)) reflects the impact of different states Z(t) on the equipment failure rate.
[0053] Weibull distribution is a statistical distribution commonly used to describe equipment life, and its probability density function is:
[0054]
[0055] Where x is the equipment failure time, β is the shape parameter, and η is the scale parameter. Based on the Weibull distribution, the probability of equipment failure within a given time, i.e., the failure rate, is calculated. The probability of equipment failure within time tt can be expressed by the cumulative distribution function F(t):
[0056]
[0057] The probability of a device failing in the time interval [t, t+Δt][t, t+Δt] can be calculated using the following formula:
[0058] P(t,Δt)=F(t+Δt)-F(t)
[0059] The probability of equipment failure in different time periods is calculated based on the equipment historical data and the parameters of the Weibull distribution, and then a PHM-based equipment failure rate model is established. By combining PHM technology and the equipment failure rate model, cigarette quality supervision and measurement instruments can better realize real-time monitoring and prediction of equipment status, which helps to reduce the failure rate of the production line and improve equipment reliability and production efficiency.
[0060] Figure 1 In, X k represents the time that the device stays in state k, assuming that the length of the stay time is random; T k represents the transition time of the kth state, where T0 = 0. Since state n is an absorbing state, X n =∞,T n =∞.
[0061] For a new device, t0 = 0, the device state at the current moment is Z(0) = 1, and the reliability function of the device is:
[0062]
[0063] Where T represents the trouble-free operation time of the equipment, h(s,Zj): represents that at time s, the equipment is in state Z j The hazard rate function h reflects the instantaneous rate of failure of the equipment under a specific state and time.
[0064] For an old device, the device has been running for t0 time, and t0≠0. Assuming that the current device state is Z(t0)=j, and the device state at time t is Z(t)=k, the reliability of the device at time t is:
[0065]
[0066] After building the data model, data generation and analysis are performed.
[0067] The state transition probability matrix can be obtained by analyzing the historical state transition data of the device. The operation status of the device in different states can be recorded, and then the number of times the device transfers to other states in each state is counted and normalized to obtain the transition probability. These probabilities constitute the state transition probability matrix.
[0068] The process of converting the number of transitions to the transition probability matrix P is as follows: The number of transitions is as follows: The number of transitions from state 1 to state 1: N11, the number of transitions from state 1 to state 2: N12, the number of transitions from state 1 to state 3: N13, the number of transitions from state 2 to state 1: N21, the number of transitions from state 2 to state 2: N22, the number of transitions from state 2 to state 3: N23, the number of transitions from state 3 to state 1: N31, the number of transitions from state 3 to state 2: N32, the number of transitions from state 3 to state 3: N33.
[0069] Each element P of the transition probability matrix P ij represents the probability of transitioning from state i to state j. The present invention can calculate the transition probability by dividing each transition number by the total transition number of the corresponding state. ij The calculation formula is as follows:
[0070]
[0071] Where n represents the total number of states, Represents the total number of transitions of state i. According to this formula, each element of the transition probability matrix P can be calculated in turn.
[0072] This paper takes the tobacco quality supervision and measurement instruments of Z Tobacco Company as the evaluation object, and summarizes the comprehensive evaluation indicators of the operation and management of tobacco quality supervision and measurement instruments by carefully analyzing and investigating the current status and problems of the operation and management of tobacco quality supervision and measurement instruments, and combining the summary and arrangement of methods such as literature analysis. The comprehensive evaluation indicators of tobacco quality supervision and measurement instruments are shown in Table 1.
[0073] According to the important test items of tobacco quality supervision, all measuring instruments for tobacco quality supervision are summarized and classified according to the importance of tobacco testing equipment. Key equipment is the equipment that should be configured for important test items, while general equipment is the equipment that should be configured or selected for more important test items. Key equipment is controlled according to the requirements of Class A monitoring and measuring equipment of each tobacco factory, and general equipment can be controlled according to the requirements of Class B or Class C monitoring and measuring equipment of each tobacco factory. This table reflects the key equipment involved in 18 important test items of tobacco quality supervision. Taking the moisture content in tobacco as an example, the key equipment is an electronic balance and an oven. The measurement codes of the oven are WCJ001, WCJ002, and WCJ003.
[0074] Table 1 Tobacco quality supervision measurement unit time interval moisture content test results of oven equipment
[0075]
[0076] According to the measurement standard of production equipment, the status equipment is divided into: Status 1: moisture content is greater than or equal to 12.75%; Status 2: moisture content is between 12.65% and 12.74%; Status 3: moisture content is less than or equal to 12.64%.
[0077] Table 2 Tobacco quality supervision measurement oven equipment moisture content detection status classification
[0078]
[0079] Generate state switching sequence: Determine the state at each time point based on the moisture content data. That is, 12.73% corresponds to state 1, 12.63% corresponds to state 3, and so on. Calculate the number of transitions based on the state sequence: Count the number of transitions between each state in the state sequence. Record the number of state transitions as the final state transition data.
[0080] Table 3 Statistics of tobacco quality supervision and measurement oven equipment detection state switching
[0081]
[0082]
[0083] 12.71% to 12.68% moved from state 1 to state 2, 12.68% to 12.64% remained in state 2, 12.64% to 12.71% moved from state 2 to state 1, 12.71% to 12.74% remained in state 1, 12.74% to 12.73% remained in state 1, 12.73% to 12.73% remained in state 1, 12.73% to 12.67% moved from state 1 to state 2 , 12.67% to 12.62% are transfers from state 2 to state 3, and 12.62% to 12.68% are transfers from state 3 to state 2. Based on the above observations, the present invention obtains the following number of transfers: number of transfers from state 1 to state 2: 2 times, number of transfers from state 1 to state 1: 2 times, number of transfers from state 2 to state 1: 1 time, number of transfers from state 2 to state 3: 1 time, number of transfers from state 3 to state 2: 1 time.
[0084] Calculate the probability of state 1 transferring to other states: the probability of transferring from state 1 to state 1 P11 = 0.5, the probability of transferring from state 1 to state 2 P12 = 0.5, and the probability of transferring from state 1 to state 3 P13 = 0. Calculate the probability of state 2 transferring to other states: the probability of transferring from state 2 to state 1 P21 = 0.5, the probability of transferring from state 2 to state 2 P22 = 0.5, and the probability of transferring from state 2 to state 3 P23 = 0, because this transfer has not been observed. Calculate the probability of state 3 transferring to other states: the probability of transferring from state 3 to state 1 P31 = 0, because this transfer has not been observed. The probability of transferring from state 3 to state 2 P32 = 1, because this transfer has been observed. The probability of transferring from state 3 to state 3 P33 = 0, because this transfer has not been observed. The transition probability matrix P is obtained as follows:
[0085]
[0086] The equipment reliability model in the present invention adopts a proportional failure model, the benchmark function uses a two-parameter Weibull distribution, and the link function uses an exponential distribution. The parameters of the model can be obtained by evaluating historical data, and the present invention will not discuss them in detail. The benchmark function and link function in the proportional probability model are as follows:
[0087]
[0088] ψ(Z(s))=e 0.5(Z(s)-1)
[0089] Among them, Z(s) represents the aging status of the equipment.
[0090] In order to verify the effectiveness of the PHM reliability evaluation method and model proposed in the present invention, the oven measurement data of 10 unit times are statistically analyzed using the model.
[0091] In order to simplify the model, the present invention divides the equipment aging state into three stages, Z(s) = {1, 2, 3}. Among them, stage 3 is the absorption state, so after the equipment enters stage 3, its residence time is considered to be: X3 = ∞. The initial distribution of the residence time of the equipment in stage 1 and stage 2 obeys the Weibull distribution, and the probability density function of the distribution is:
[0092]
[0093] The parameters α and β of the Weibull distribution are estimated using maximum likelihood estimation (MLE).
[0094] The maximum likelihood function is:
[0095]
[0096] Taking the logarithm gives the log-likelihood function:
[0097]
[0098] By taking the derivative of the log-likelihood function with respect to α and β and setting them to zero, we can obtain the parameter estimates, α=1.1077, β=1.5. Using the estimated parameters, we calculate the failure rate function h(t) of the Weibull distribution.
[0099]
[0100] For a device that has been running for t0=5 and is currently in state Z(t0)=2, the state at time t=8 is Z(t)=3, and its reliability function is:
[0101]
[0102] The reliability function can be used to estimate the probability of no failure after a unit of time. For example, R(10) represents the probability that the equipment is still fault-free after running for 10 time units, and R(20) represents the probability that the equipment is still fault-free after running for 20 time units. The reliability function can be used to predict the probability of a device running without failure within a specific time period, thereby estimating the life of the equipment. This helps plan maintenance and replacement of equipment, reducing downtime and economic losses caused by equipment failure.
[0103] The above is only a preferred embodiment 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 principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.
Claims
1. A reliability evaluation model construction method for cigarette quality supervision and measurement system. Features: The two-parameter Weibull distribution is used as the benchmark function and the exponential distribution is used as the link function to establish the PHM equipment failure rate model. The failure rate and life distribution of the equipment are estimated by analyzing the historical failure data and current status data of the equipment.
2. The reliability evaluation model construction method for the cigarette quality supervision and measurement system according to claim 1, Features: The specific forms of the benchmark function and the connection function are as follows: h(t,Z(t))=h 0 (t)ψ(Z(t)),t≥0 Where: t is the total operating time of the equipment after commissioning; Z(t) is the state of the equipment at time t; the benchmark failure rate function h 0 (t) is used to represent the aging process of the equipment, and Weibull distribution is selected; the connection function ψ(Z(t)) reflects the impact of different states Z(t) on the equipment failure rate.
3. The reliability evaluation model construction method for a cigarette quality monitoring and measurement system according to claim 1, Features: Weibull distribution is a statistical distribution used to describe the life of equipment. Its probability density function is: Where x is the equipment failure time, β is the shape parameter, and η is the scale parameter. Based on the Weibull distribution, the probability of equipment failure within a given time is calculated, which is recorded as the failure rate.
4. The reliability evaluation model construction method for a cigarette quality monitoring and measurement system according to claim 1, Features: The probability of a device failing within time tt can be expressed using the cumulative distribution function F(t): The probability of a device failing in the time interval [t, t+Δt][t, t+Δt] can be calculated using the following formula: P(t,Δt)=F(t+Δt)-F(t).
5. The reliability evaluation model construction method for a cigarette quality monitoring and measurement system according to claim 1, Features: For a new device, t 0 =0, the current state of the device is Z(0)=1, and the reliability function of the device is: Where T represents the trouble-free operation time of the equipment, h(s,Zj): represents that at time s, the equipment is in state Z j The hazard rate function h reflects the instantaneous rate of failure of the equipment under a certain state and time. For an old device, the device is currently running t 0 time, and t 0 ≠0, assuming that the current device state is Z(t 0 )=j, the state of the device at time t is Z(t)=k, then the reliability of the device at time t is: Among them, X k represents the time the device stays in state k, assuming that the length of the stay time is random; Tk represents the transition time of the kth state, Z j It means that the device is at time t 0 The state at the moment, ZK represents the state of the device at time t.
6. The reliability evaluation model construction method for a cigarette quality monitoring and measurement system according to claim 1, Features: After the data model is established, data generation and analysis are performed. The state transition probability matrix is obtained by analyzing the historical state transition data of the equipment. The operation status of the equipment in different states is recorded, and then the number of times the equipment transfers to other states in each state is counted and normalized to obtain the transition probability. The above transition probabilities constitute the state transition probability matrix.
7. The reliability evaluation model construction method for a cigarette quality monitoring and measurement system according to claim 6, Features: Each element P of the transition probability matrix P ij represents the probability of transitioning from state i to state j. The transition probability is calculated by dividing each transition count by the total transition count of the corresponding state. The transition probability P ij The calculation formula is as follows: Where n represents the total number of states, Represents the total number of transitions of state i. According to this formula, each element of the transition probability matrix P is calculated in turn.
8. The reliability evaluation model construction method for a cigarette quality monitoring and measurement system according to claim 1, Features: The equipment reliability model adopts the proportional failure model, the benchmark function uses the two-parameter Weibull distribution, the connection function uses the exponential distribution, and the parameters of the model are obtained through historical data evaluation. The benchmark function h in the proportional probability model 0 (s) and the connection function ψ(Z(s)) are as follows: ψ(Z(s))=e 0.5(Z(s)-1) Among them, Z(s) represents the aging status of the equipment.
