Reliability, Availability, Maintainability and Survivability

Introduction

The Reliability, Availability, Maintainability and Survivability (RAMS) module is a software that assesses reliability, availability, maintainability, and survivability of marine energy conversion systems from the early concept stage to the commercial deployment. The definitions of reliability, availability, maintainability, and survivability, are given below:

reliabiity is the ability of a structure or structural member to fulfil the specified requirements, during the working life, for which it has been designed.
availability theoretically refers to the probability that a system or component is performing its required function at a given point in time or over a stated period of time when operated and maintained in a prescribed manner. In most of the engineering applications, it is defined as the ratio of the uptime to the design lifetime.
maintainability is the ability of a system to be repaired and restored to service when maintenance is conducted by personnel using specified skill levels and prescribed procedures and resources.
survivability is the probability that the converter will stay on station over the stated operational life.

Structure

This module’s documentation is divided into four main sections:

Tutorials to give step-by-step instructions on using RAMS for new users.
How-to Guides that show how to achieve specific outcomes using RAMS.
A section on background, theory and calculation methods that describes how RAMS works and aims to give confidence in the tools.
The API reference section documents the code of modules, classes, API, and GUI.

Functionalities

The Reliability, Availability, Maintainability and Survivability module has six major functionalities:

reliability assessment– contains both component-level and system-level assessments. The component-level reliability assessment simulates the time to failure (TTF) of basic components; estimates the mean values, maximum and standard deviations of TTFs of basic components. The system-level reliability assessment simulates the time to failure (TTF) of subsystems (energy delivey, energy transformation and station keeping) and the array; estimates the mean values, maximum and standard deviations of TTFs of these subsystems and the array; calculates the maximum annual probability of failure (PoF) of these subsystems and the array.
availability assessment– calculates the availability of the individual devices and the average availability of the array.
maintainability assessment– calculates the probability that the damaged components can be repaired within a specific period of time, given presrivbed resources and equipment.

Workflow for using the RAMS module

The four features, namely reliability, availability, maintainability and survivability, are assessed separately in the RAMS module. The generic workflows are the same, which includes collection of inputs, check the inputs, perform assessment and view the results.

Figure 1 Workflow of RAMS Module -Reliability assessment

Figure 2 Workflow of RAMS Module - Availability assessment

Figure 3 Workflow of RAMS Module - Maintainability assessment

Figure 4 Workflow of RAMS Module - Survivability assessment

Overview of RAMS data requirements

Reliability assessment requires the hierarchies of the ED, ET and SK subsystems, the number of simulations and the waiting time, as summarized in the following table.

Table 1 Summary of Inputs for Reliability Assessment

External module inputs	Default	Data origin	Units
ED hierarchy	Required	ED or user-defined
ET hierarchy	Required	ET or user-defined
SK hierarchy	Required	SK or user-defined
Number of simulations	Required	User-defined
Waiting time	Required	User-defined	hour

Availability assessment requires the downtime of all the devices in an array, as summarized in the following table.

Table 2 Summary of Inputs for Availability Assessment

External module inputs	Default	Data origin	Units
Downtime	Required	LMO or user-defined

Maintainability assessment requires the downtime of all the devices in an array, as summarized in the following table.

Table 3 Summary of Inputs for Maintainability Assessment

External module inputs	Default	Data origin	Units
Available time	Required	User-defined	hour
Probability distribution of repair time	Required	User-defined
Standard deviation of repair time	Required	SK or user-defined	hour
MTTR	Required	LMO or user-defined
Technologies	Required	LMO or user-defined

Survivability assessment requires the inputs in the following table.

Table 4 Summary of Inputs for Survivability Assessment

External module inputs	Default	Data origin
stress_sk.json	Required	SK or user-defined
stress_et.json	Required	ET or user-defined
Other parameters	Required	Default or user-defined

The following table summarizes the other parameters and the explanations.

Table 5 Explanation of Other Parameters for Survivability Assessment

Parameters	Format	Explanation
cov_a	float	The coefficient of variance of the S-N curve parameter a
cov_l	float	The coefficient of variance of the extreme/ ultimate load
cov_q	float	See Note 1)
cov_r	float	See Note 2)
cov_ufl	float	The coefficient of variance of the uncertainty factor associated with the load\|
cov_ufr	float	See Note 3)
mu_ufl	float	The mean value of the uncertainty factor associated with the load
mu_ufr	float	The mean value of the uncertainty factor associated with the resistance
n_sim_fls	integer	The number of simulations for the survivability assessment (fatigue limit state, FLS)
n_sim_uls	integer	The number of simulations for the survivability assessment (ultimate limit state, ULS)
option_fls	string	See Note 4)
option_uls	string	See Note 5)
pd_a	string	The probability distribution of the S-N curve parameter a
pd_h	string	See Note 6)
pd_l	string	The probability distribution of the load
pd_m	string	The probability distribution of the S-N curve parameter m
pd_n	string	The probability distribution of the number of stress range cycles
pd_q	string	See Note 7)
pd_r	string	The probability distribution of the resistance
pd_ufl	string	The probability distribution of the uncertainty factor associated with the load
pd_ufr	string	The probability distribution of the uncertainty factor associated with the resistance

Notes:

1) The coefficient of variance of the scale parameter of the 2-parameter Weibull distribution (assumed that the long-term stress ranges follow the 2-parameter Weibull distribution).

2) The coefficient of variance of the resistance (maximum breaking load, MBL) of the mooring lines.

3) The coefficient of variance of the uncertainty factor associated with the resistance.

4) The method used for assessing the survivability (FLS), option 1 – ‘Monte Carlo’ (for complexity 1, 2 & 3); option 2 – ‘FORM’ (for complexity 2 & 3).

5) The method used for assessing the survivability (ULS), option 1 – ‘Monte Carlo’ (for complexity 1, 2 & 3); option 2 – ‘FORM’ (for complexity 2 & 3).

6) The probability distribution of the shape parameter of the 2-parameter Weibull distribution (assumed that the long-term stress ranges follow the 2-parameter Weibull distribution).

7) The probability distribution of the scale parameter of the 2-parameter Weibull distribution (assumed that the long-term stress ranges follow the 2-parameter Weibull distribution).

The data structure of stress_sk.json and stress_et.json are described as follows:

Table 6 Explanation of Data in stress_sk.json

Data	Data origin
The ultimate loads on the mooring lines	devices[i][“uls_results”][“mooring_tension”]
The maximum breaking loads (MBL) of the mooring lines	devices[i][“uls_results”][“mbl_uls”]
The stress ranges on the mooring lines	devices[i][“fls_results”][“cdf_stress_range”]
The cumulative distribution functions (CDFs) of the stress ranges	devices[i][“fls_results”][“cdf”]
The S-N curve parameter a	devices[i][“fls_results”][“ad”]
The S-N curve parameter m	devices[i][“fls_results”][“m”]
The number of stress range cycles	devices[i][“fls_results”][“n_cycles_lifetime”]

Notes: 1) the key “devices” is inlcuded in the raw output json file from the SK module.

Table 7 Explanation of Data in stress_et.json

Data	Data origin
The S-N curve	[i][“S_N”]
The ultimate stresses	[i][“ultimate_stress”]
The maximum stresses and probability	[i][“maximum_stress_probability”]
The fatigue stresses and probability	[i][“fatigue_stress_probability”]
The number of cycles of stress ranges	[i][“number_cycles”]

Notes:

1) “S_N” is a dictionary, including four kesys, namely, “description”, “lavel”, “unit” and “value”.

The key “value” contains the actual data, which is a 2D list. Each 1D list contains the S-N curve paramers. The first and second 1D lists contain the parameters corresponding to the first and second parts of a bilinear S-N curve. For example, [“value”][0] = [3.0, 10.97]. m=3.0 log10(a)=10.97.

2) “ultimate_stress” is a dictionary, including four kesys, namely, “description”, “lavel”, “unit” and “value”. The key “value” contains the actual data, which is a float.

3) “maximum_stress_probability” is a dictionary, including four kesys, namely, “description”, “lavel”, “unit” and “value”. The key “value” contains the actual data, namely “stress” and “proability”. Both “stress” and “probability” contain 1D list.

4) “fatigue_stress_probability” is a dictionary, including four kesys, namely, “description”, “lavel”, “unit” and “value”. The key “value” contains the actual data, namely “stress” and “proability”. Both “stress” and “probability” contain 1D list.

5) “number_cycles” is a dictionary, including four kesys, namely, “description”, “lavel”, “unit” and “value”. The key “value” contains the actual data, which is a 1D list. It should be stressed that the each entry must correspond to the correspnding stress range in the key “value” in “fatigue_stress_probability”.