A method for formulating variable maintenance cycles for urban rail electrical equipment considering environmental impact
Through K-S inspection and neural network technology, variable maintenance cycles of urban rail transit electrical equipment are formulated to consider environmental factors, solving the problem of environmental factors not being considered in the existing technology, and improving the level of equipment management and operational safety.
Patent Information
- Application Number
- CN202111255861.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-10-27
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2041-10-27
AI Technical Summary
The prior art does not consider environmental factors when determining the maintenance cycle of urban rail transit electrical equipment, resulting in frequent equipment failures and increases operating costs and safety risks.
The K-S test method is used to select the equipment failure probability distribution model, and the neural network is used to construct the relationship between environmental variables and distribution parameters, and a variable maintenance period based on reliability is formulated.
Through the optimization method, the probability of equipment failure is reduced, scientific maintenance plans are provided, and equipment management level and operational safety are improved.
Smart Images

Figure CN114091573B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of maintenance management of urban rail transit equipment, and in particular to a method for formulating a variable maintenance cycle of urban rail electrical equipment taking environmental impact into consideration. Background Art
[0002] The stable operation of various equipment in the urban rail transit system is the premise for ensuring operational safety and quality. For urban rail transit enterprises, frequent equipment failures will increase the workload of maintenance personnel and increase the total cost of equipment maintenance. Sudden equipment failures may not only cause equipment downtime, reduce equipment reliability, reduce production efficiency and service levels, but also endanger the lives and property safety of passengers. In 2009, the Washington Metro Red Line vehicle collision accident caused 9 deaths and 80 injuries. The cause of the accident was a signal equipment failure; in 2014, the Seoul Metro Line 2 rear-end collision accident caused more than 240 injuries. The cause of the accident was abnormal on-board equipment of the front vehicle; in 2014, the Moscow Metro derailment accident caused 23 deaths and 161 injuries. The cause of the accident was a turnout equipment failure. Practice has proved that the direct cause of major accidents in urban rail transit systems is often equipment failure in addition to deliberate human damage. These vivid lessons are everywhere, and they have also sounded the alarm for us, reminding us to pay more attention to equipment maintenance and improve the level of equipment maintenance management.
[0003] Maintenance is a technical activity carried out to maintain and restore the equipment to good working condition. After a period of time, the failure patterns of urban rail transit equipment will change due to environmental factors. Adopting a fixed-cycle maintenance strategy will increase the number of sudden equipment shutdowns. Therefore, the maintenance cycle of the equipment should be adjusted accordingly according to the changes in influencing factors. At present, the maintenance cycle of urban rail transit equipment is mostly formulated with reference to national railway specifications and maintenance experience in the industry, without considering the impact of environmental factors on the operating conditions of the equipment. Especially electrical equipment, which does not have the stable aging process of mechanical equipment. In different seasons, the failure conditions of electrical equipment are significantly different. Therefore, environmental factors should be considered to formulate corresponding variable-cycle maintenance cycles for urban rail electrical equipment.
[0004] Scholars at home and abroad have proposed several methods to determine the equipment maintenance cycle. Among them, the method of modeling and solving the maintenance cycle with the maximum availability and the minimum average maintenance cost as the goal is the most widely used. However, these studies did not consider the impact of environmental factors on the equipment operation status when fitting the probability distribution. Some scholars introduced the service age reduction factor and the failure rate increase factor to calculate the hybrid failure rate and apply it to the equipment maintenance cycle model. It takes into account the impact of the number of failures, but it is difficult to apply to electrical equipment that is affected by environmental changes and has relatively random failures. Therefore, in order to reasonably save the equipment maintenance resources of urban rail transit operating units and improve the level of equipment maintenance management, there is an urgent need for a method for formulating variable maintenance cycles for urban rail transit electrical equipment that is highly practical and can follow changes in environmental influences. Summary of the invention
[0005] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a method for formulating a variable maintenance period for urban rail electrical equipment taking into account environmental influences. The method selects the optimal fitting distribution function of the equipment failure probability distribution model based on the KS test method and estimates the distribution parameters using the maximum likelihood estimation method to obtain the equipment failure probability distribution model under different environmental influences; obtains the relationship between environmental variables and distribution parameters based on a neural network, and constructs a conditional probability distribution model of equipment failure under certain environmental influences; and finally formulates a variable maintenance period for urban rail electrical equipment with reliability as a constraint.
[0006] The present invention adopts the following technical solutions to solve the above technical problems:
[0007] A method for formulating a variable maintenance cycle for urban rail electrical equipment considering environmental impacts according to the present invention comprises the following steps:
[0008] Step S1, collecting and storing the fault data and environmental data of the device, and saving the fault data and environmental data of the device in the device database; wherein the fault data includes the name of the fault module of the device, the fault time, the fault phenomenon, the fault impact and the fault repair time, and the environmental data includes the temperature data and the humidity data;
[0009] Step S2, calculating the equipment trouble-free operation time data to obtain the equipment trouble-free operation time data set, and adding the equipment trouble-free operation time data set to the equipment database;
[0010] Step S3: Select the optimal fitting distribution function of the equipment failure probability distribution model based on the equipment trouble-free operation time data and the KS test method. The specific steps are as follows:
[0011] S31, establishing equipment failure probability distribution models using exponential distribution function, lognormal distribution function and Weibull distribution function respectively;
[0012] S32, estimating the parameters of each distribution function in S31 using the equipment trouble-free operation time data and the maximum likelihood estimation method;
[0013] S33. Use the KS test method to test the goodness of fit of each distribution function to the equipment trouble-free operation time data, and select the test statistic D max The distribution function corresponding to the minimum is taken as the optimal fitting distribution function of the equipment failure probability distribution model; where the test statistic D max The calculation formula is as follows:
[0014]
[0015] Among them, t i is the trouble-free operation time of the equipment when it fails for the i-th time, F(t i ) is the unreliability of the equipment at the time of the i-th failure, F^(t i ) is the observed value of median rank unreliability of the equipment at the time of the i-th failure, and n is the total number of equipment failures;
[0016] Step S4, constructing a neural network-based equipment failure condition probability distribution model according to the data stored in the equipment database and the optimal fitting distribution function selected in S33;
[0017] Step S5: Taking reliability as a constraint condition, the variable maintenance period of the equipment is calculated according to the equipment failure condition probability distribution model.
[0018] As a further optimization scheme of the method for formulating a variable maintenance cycle of urban rail electrical equipment considering environmental impacts according to the present invention, the specific steps of step S4 are as follows:
[0019] S41. Using the fault module name and fault time as keywords, query the temperature data, humidity data, and equipment trouble-free operation time data of the equipment in the equipment database to obtain the equipment fault sample data set U = {(t1, w1, s1), (t2, w2, s2), ..., (t n ,w n ,s n )}, where w i is the temperature of the surrounding environment when the equipment fails for the i-th time, s i is the humidity of the surrounding environment when the equipment fails for the i-th time, 1≤i≤n; randomly shuffle the data in U and divide the shuffled U into the training set U train and the test set U test ;
[0020] S42, U train The minimum and maximum values of the medium temperature are denoted as w min 、w max , U trainThe maximum and minimum values of humidity are denoted as s max 、s min ; respectively divide the interval [w min ,w max ]、[s min ,s max ] is divided into k1 and k2 small intervals, where k1 and k2 are positive integers; the training set U train is divided into k1*k2 mutually exclusive sub-datasets, U train The relationship between each sub-dataset is as follows:
[0021] U train =U1∪U2∪......∪U k1*k2 (2)
[0022] Among them, U j is the training set U train The sub-dataset under the influence of the j-th environment, j = 1, 2, ..., k1 * k2;
[0023] S43, constructing k1*k2 equipment failure probability distribution models under different environmental influences according to the optimal fitting distribution function selected in S33, and the equipment failure probability distribution model under the jth environmental influence Among them, F(t,γ j ) is the unreliability function of the equipment, t is U j The equipment trouble-free operation time data in is the distribution parameter of the optimal fitting distribution function under the influence of the jth environment, γ k,j For γ j The kth distribution parameter in, k = 1, 2, ... z j , z j is the total number of distribution parameters of the optimal fitting distribution function under the influence of the jth environment; based on U j The equipment failure-free operation time data t and γ in j , we can get the equipment failure probability distribution model F under the jth environmental influence j ;
[0024] S44, respectively, with U j The average temperature in and humidity average Represents U j The temperature and humidity levels, combined with γ j , build a neural network model based on the attention model; As the input vector x of the neural network model, [γ1,γ2,......,γ k1*k2] is used as the output vector y of the neural network model, and the nonlinear mapping relationship between the environmental variables and the distribution parameters of the equipment failure probability distribution model is obtained: The environmental variables include the average temperature and humidity average is the functional expression of the nonlinear mapping relationship; Substitute F j Obtain the equipment failure probability distribution model under the influence of the jth environment For t, The equipment unreliability function is used to obtain the equipment failure conditional probability distribution model F under certain environmental influences. The expression is as follows:
[0025]
[0026] As a further optimization scheme of the method for formulating a variable maintenance cycle of urban rail electrical equipment considering environmental impacts according to the present invention, the specific steps of step S5 are as follows:
[0027] S51. Calculate the equipment reliability function R, the formula is as follows:
[0028]
[0029] in, For t, Equipment reliability function;
[0030] S52. Substitute the environmental data and the minimum reliability R0 required for equipment operation into formula (4) to obtain the variable maintenance cycle of the equipment under different environmental influences.
[0031] As a further optimization scheme of the method for formulating a variable maintenance cycle of urban rail electrical equipment considering environmental impact described in the present invention, in S43, the maximum likelihood estimation method is used to estimate γ j .
[0032] As a further optimization scheme of the method for formulating variable maintenance cycles for urban rail electrical equipment taking into account environmental impacts described in the present invention, the equipment trouble-free operation time data in step S2 refers to the time interval between the current fault time of the equipment and the last fault repair time, in units of hours or days.
[0033] As a further optimization scheme of the method for formulating variable maintenance periods of urban rail electrical equipment considering environmental impacts described in the present invention, the neural network model based on the attention model in step S44 has a 4-layer structure: the first layer is the input layer, and the number of neurons is equal to the number of input environmental variables; the second and third layers are hidden layers, respectively containing 32 and 64 neurons; the fourth layer is the output layer, and the number of neurons is consistent with the number of parameters in the equipment failure probability distribution model; a SEblock lightweight attention model is connected in parallel between the input layer and the second layer, which calculates the input parameter weight information through two fully connected layers; and a Mish activation function is selected between layers.
[0034] As a further optimization scheme of the method for formulating variable maintenance cycle of urban rail electrical equipment considering environmental impact described in the present invention, the test set U test The data is used to evaluate and analyze the benefits of the variable maintenance cycle of the equipment.
[0035] As a further optimization scheme of the method for formulating variable maintenance cycle of urban rail electrical equipment considering environmental impact described in the present invention, the test set U test The equipment trouble-free operation time data is divided by season, and it is believed that faults that are greater than the corresponding maintenance cycle of each season will be detected and eliminated before equipment failure.
[0036] Compared with the prior art, the present invention adopts the above technical solution and has the following technical effects:
[0037] The present invention tests the goodness of fit of each distribution function through the KS test method, and selects the best fitting distribution function for the equipment failure probability distribution model; on the basis of the equipment failure probability distribution model, the relationship between environmental variables and distribution parameters is obtained based on a neural network, and a conditional probability distribution model of equipment failure under certain environmental influences is further obtained; a variable maintenance cycle of urban rail electrical equipment is formulated with reliability as a constraint condition, which is more in line with the actual operation of the equipment under environmental influences, reduces the probability of sudden failure of the equipment, and provides a certain scientific basis for rail transit operating units to formulate equipment maintenance plans. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] Figure 1 It is a specific flow chart of the method of the present invention;
[0039] Figure 2 Schematic diagram of the neural network structure of the present invention. DETAILED DESCRIPTION
[0040] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.
[0041] like Figure 1The figure shows a specific process of a method for formulating a variable maintenance cycle for urban rail electrical equipment taking environmental impact into consideration. The present invention is further described below in conjunction with an embodiment.
[0042] Step S1, collecting and storing the fault data and environmental data of the device, and saving the fault data and environmental data of the device in the device database; wherein the fault data includes the name of the fault module of the device, the fault time, the fault phenomenon, the fault impact and the fault repair time, and the environmental data includes the temperature data and the humidity data;
[0043] Step S2, calculating the equipment trouble-free operation time data to obtain the equipment trouble-free operation time data set, and adding the equipment trouble-free operation time data set to the equipment database;
[0044] Step S3: Select the optimal fitting distribution function of the equipment failure probability distribution model based on the equipment trouble-free operation time data and the KS test method. The specific steps are as follows:
[0045] S31, establishing equipment failure probability distribution models using exponential distribution function, lognormal distribution function and Weibull distribution function respectively;
[0046] S32, estimating the parameters of each distribution function in S31 using the equipment trouble-free operation time data and the maximum likelihood estimation method;
[0047] S33. Use the KS test method to test the goodness of fit of each distribution function to the equipment trouble-free operation time data, and select the test statistic D max The distribution function corresponding to the minimum is taken as the optimal fitting distribution function of the equipment failure probability distribution model; where the test statistic D max The calculation formula is as follows:
[0048]
[0049] Among them, t i is the trouble-free operation time of the equipment when it fails for the i-th time, F(t i ) is the unreliability of the equipment at the time of the i-th failure, F^(t i ) is the observed value of median rank unreliability of the equipment at the time of the i-th failure, and n is the total number of equipment failures;
[0050] Step S4, constructing a neural network-based equipment failure condition probability distribution model according to the data stored in the equipment database and the optimal fitting distribution function selected in S33;
[0051] Step S5: Taking reliability as a constraint condition, the variable maintenance period of the equipment is calculated according to the equipment failure condition probability distribution model.
[0052] The specific steps of step S4 are as follows:
[0053] S41. Using the fault module name and fault time as keywords, query the temperature data, humidity data, and equipment trouble-free operation time data of the equipment in the equipment database to obtain the equipment fault sample data set U = {(t1, w1, s1), (t2, w2, s2), ..., (t n ,w n ,s n )}, where w i is the temperature of the surrounding environment when the equipment fails for the i-th time, s i is the humidity of the surrounding environment when the equipment fails for the i-th time, 1≤i≤n; randomly shuffle the data in U and divide the shuffled U into the training set U train and the test set U test ;
[0054] S42, U train The minimum and maximum values of the medium temperature are denoted as w min 、w max , U train The maximum and minimum values of humidity are denoted as s max 、s min ; respectively divide the interval [w min ,w max ]、[s min ,s max ] is divided into k1 and k2 small intervals, where k1 and k2 are positive integers; the training set U train is divided into k1*k2 mutually exclusive sub-datasets, U train The relationship between each sub-dataset is as follows:
[0055] U train =U1∪U2∪......∪U k1*k2 (6)
[0056] Among them, U j is the training set U train The sub-dataset under the influence of the j-th environment, j = 1, 2, ..., k1 * k2;
[0057] S43, constructing k1*k2 equipment failure probability distribution models under different environmental influences according to the optimal fitting distribution function selected in S33, and the equipment failure probability distribution model under the jth environmental influence Among them, F(t,γ j ) is the unreliability function of the equipment, t is U j The equipment trouble-free operation time data in is the distribution parameter of the optimal fitting distribution function under the influence of the jth environment, γ k,j For γ jThe kth distribution parameter in, k = 1, 2, ... z j , z j is the total number of distribution parameters of the optimal fitting distribution function under the influence of the jth environment; based on U j The equipment failure-free operation time data t and γ in j , we can get the equipment failure probability distribution model F under the jth environmental influence j ;
[0058] S44, respectively, with U j The average temperature in and humidity average Represents U j The temperature and humidity levels, combined with γ j , construct a neural network model based on the attention model. The network structure diagram of the neural network model is as follows Figure 2 shown; As the input vector x of the neural network model, [γ1,γ2,......,γ k1*k2 ] is used as the output vector y of the neural network model, and the nonlinear mapping relationship between the environmental variables and the distribution parameters of the equipment failure probability distribution model is obtained: The environmental variables include the average temperature and humidity average is the functional expression of the nonlinear mapping relationship; Substitute F j Obtain the equipment failure probability distribution model under the influence of the jth environment For t, The equipment unreliability function is used to obtain the equipment failure conditional probability distribution model F under certain environmental influences. The expression is as follows:
[0059]
[0060] The specific steps of step S5 are as follows:
[0061] S51. Calculate the equipment reliability function R, the formula is as follows:
[0062]
[0063] in, For t, Equipment reliability function;
[0064] S52. Substitute the environmental data and the minimum reliability R0 required for equipment operation into formula (4) to obtain the variable maintenance cycle of the equipment under different environmental influences.
[0065] In step S43, the maximum likelihood estimation method is used to estimate γ j .
[0066] The equipment trouble-free operation time data in step S2 refers to the time interval between the current equipment trouble time and the last trouble repair time, in hours or days.
[0067] The neural network model based on the attention model in step S44 has a four-layer structure: the first layer is the input layer, and the number of neurons is equal to the number of input environmental variables; the second and third layers are hidden layers, containing 32 and 64 neurons respectively; the fourth layer is the output layer, and the number of neurons is consistent with the number of parameters in the equipment failure probability distribution model; the SEblock lightweight attention model is connected in parallel between the input layer and the second layer, which calculates the input parameter weight information through two fully connected layers; the Mish activation function is selected between layers.
[0068] Using the test set U test The data is used to evaluate and analyze the benefits of the variable maintenance cycle of the equipment.
[0069] Using the test set U test The data is used to evaluate and analyze the benefits of the variable maintenance cycle of the equipment. test The data is used to evaluate and analyze the benefits of the variable maintenance cycle of the equipment.
[0070] The test set U test The equipment trouble-free operation time data is divided by season, and it is believed that faults that are greater than the corresponding maintenance cycle of each season will be detected and eliminated before equipment failure.
[0071] The above description is only a specific implementation of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by a technician familiar with the technical field within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention.
Claims
1. A method for formulating variable maintenance cycles for urban rail electrical equipment considering environmental impacts, characterized in that: The following steps are involved: Step S1, collecting and storing the fault data and environmental data of the device, and saving the fault data and environmental data of the device in the device database; wherein the fault data includes the name of the fault module of the device, the fault time, the fault phenomenon, the fault impact and the fault repair time, and the environmental data includes the temperature data and the humidity data; Step S2, calculating the equipment trouble-free operation time data to obtain the equipment trouble-free operation time data set, and adding the equipment trouble-free operation time data set to the equipment database; Step S3: Select the optimal fitting distribution function of the equipment failure probability distribution model based on the equipment trouble-free operation time data and the KS test method. The specific steps are as follows: S31, establishing equipment failure probability distribution models using exponential distribution function, lognormal distribution function and Weibull distribution function respectively; S32, estimating the parameters of each distribution function in S31 using the equipment trouble-free operation time data and the maximum likelihood estimation method; S33. Use the KS test method to test the goodness of fit of each distribution function to the equipment trouble-free operation time data, and select the test statistic D max The distribution function corresponding to the minimum is taken as the optimal fitting distribution function of the equipment failure probability distribution model; where the test statistic D max The calculation formula is as follows: Among them, t i is the trouble-free operation time of the equipment when it fails for the i-th time, F(t i ) is the unreliability of the equipment at the time of the i-th failure, F^(t i ) is the observed value of median rank unreliability of the equipment at the time of the i-th failure, and n is the total number of equipment failures; Step S4, constructing a neural network-based equipment failure condition probability distribution model according to the data stored in the equipment database and the optimal fitting distribution function selected in S33; Step S5: Taking reliability as a constraint condition, the variable maintenance period of the equipment is calculated according to the equipment failure condition probability distribution model.
2. A method for formulating variable maintenance cycles for urban rail electrical equipment considering environmental impacts according to claim 1, characterized in that: The specific steps of step S4 are as follows: S41. Using the fault module name and fault time as keywords, query the temperature data, humidity data, and equipment trouble-free operation time data of the equipment in the equipment database to obtain the equipment fault sample data set U = {(t1, w1, s1), (t2, w2, s2), ..., (t n ,w n ,s n )}, where w i is the temperature of the surrounding environment when the equipment fails for the i-th time, s i is the humidity of the surrounding environment when the equipment fails for the i-th time, 1≤i≤n; randomly shuffle the data in U and divide the shuffled U into the training set U train and the test set U test ; S42, U train The minimum and maximum values of the medium temperature are denoted as w min 、w max , U train The maximum and minimum values of humidity are denoted as s max 、s min ; respectively divide the interval [w min ,w max ]、[s min ,s max ] is divided into k1 and k2 small intervals, where k1 and k2 are positive integers; the training set U train is divided into k1*k2 mutually exclusive sub-datasets, U train The relationship between each sub-dataset is as follows: IN train =U1∪U2∪......∪U k1*k2 (2) Among them, U j is the training set U train The sub-dataset under the influence of the j-th environment, j = 1, 2, ..., k1 * k2; S43, constructing k1*k2 equipment failure probability distribution models under different environmental influences according to the optimal fitting distribution function selected in S33, and the equipment failure probability distribution model under the jth environmental influence Among them, F(t,γ j ) is the unreliability function of the equipment, t is U j The equipment trouble-free operation time data in is the distribution parameter of the optimal fitting distribution function under the influence of the jth environment, γ k,j For γ j The kth distribution parameter in, k = 1, 2, ... z j , z j is the total number of distribution parameters of the optimal fitting distribution function under the influence of the jth environment; based on U j The equipment failure-free operation time data t and γ in j , we can get the equipment failure probability distribution model F under the jth environmental influence j ; S44, respectively, with U j The average temperature in and humidity average Represents U j The temperature and humidity levels, combined with γ j , build a neural network model based on the attention model; As the input vector x of the neural network model, [γ1,γ2,......,γ k1*k2 ] is used as the output vector y of the neural network model, and the nonlinear mapping relationship between the environmental variables and the distribution parameters of the equipment failure probability distribution model is obtained: The environmental variables include the average temperature and humidity average is the functional expression of the nonlinear mapping relationship; Substitute F j Obtain the equipment failure probability distribution model under the influence of the jth environment For t, The equipment unreliability function is used to obtain the equipment failure conditional probability distribution model F under certain environmental influences. The expression is as follows:
3. A method for formulating variable maintenance cycles for urban rail electrical equipment considering environmental impacts according to claim 2, characterized in that: The specific steps of step S5 are as follows: S51. Calculate the equipment reliability function R, the formula is as follows: in, For t, Equipment reliability function; S52. Substitute the environmental data and the minimum reliability R0 required for equipment operation into formula (4) to obtain the variable maintenance cycle of the equipment under different environmental influences.
4. A method for formulating variable maintenance cycles for urban rail electrical equipment considering environmental impacts according to claim 2, characterized in that: In S43, the maximum likelihood estimation method is used to estimate γ j .
5. A method for formulating variable maintenance cycles for urban rail electrical equipment considering environmental impacts according to claim 1, characterized in that: The equipment trouble-free operation time data in step S2 refers to the time interval between the current equipment trouble time and the last trouble repair time, in hours or days.
6. A method for formulating variable maintenance cycles for urban rail electrical equipment considering environmental impacts according to claim 2, characterized in that: The neural network model based on the attention model in step S44 has a 4-layer structure: the first layer is the input layer, and the number of neurons is equal to the number of input environmental variables; the second and third layers are hidden layers, which contain 32 and 64 neurons respectively; the fourth layer is the output layer, and the number of neurons is consistent with the number of parameters in the equipment failure probability distribution model; the SEblock lightweight attention model is connected in parallel between the input layer and the second layer, which calculates the input parameter weight information through two fully connected layers; The Mish activation function is selected between layers.
7. A method for formulating variable maintenance cycles for urban rail electrical equipment considering environmental impacts according to claim 2, characterized in that: Using the test set U test The data is used to evaluate and analyze the benefits of the variable maintenance cycle of the equipment.
8. A method for formulating variable maintenance cycles for urban rail electrical equipment considering environmental impacts according to claim 2, characterized in that: The test set U test The equipment trouble-free operation time data is divided by season, and it is believed that faults that are greater than the corresponding maintenance cycle of each season will be detected and eliminated before equipment failure.
